File size: 171,033 Bytes
077605d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
[
  {
    "id": "IF01",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: The deliverable file artifact/report.pptx exists and can be opened. Verification method: Check that artifact/report.pptx exists, and verify it using compatible rendering tools and PPTX structure parsing. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 11: Deliver artifact/report.pptx.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G01"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF02",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: The PPTX has a total of six slides. Verification method: Open artifact/report.pptx and count the total number of slides as 6. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 3: create a six-slide English PPTX.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G02"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF03",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: The main body of the PPTX is in English. Verification method: Extract the titles, body text, and figure captions page by page, and confirm that the main body is in English; formulas and proper nouns are excepted. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 3: English PPTX.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G03"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF04",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: The six pages must include a cover page. Verification method: Locate the cover page and confirm that it carries the title/report information. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 3: including cover/conclusion.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G04"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF05",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: The six pages must include a conclusion page. Verification method: Locate the conclusion page and confirm that it carries the conclusion/boundary summary. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 3: including cover/conclusion.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G04"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF06",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: Must cover the research question. Verification method: Locate the page and text that state the research question and check them. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 5: Cover the research question.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G05"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF07",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: Must cover the Transformer–MPC connection. Verification method: Locate the corresponding page and check whether it states the connection between Transformer and MPC. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 5: the Transformer–MPC connection.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G05"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF08",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: Must cover the k-hop task and its relationship to depth. Verification method: Locate the corresponding page and check the statement of the k-hop task and depth relationship. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 5: the k-hop task and depth relationship.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G05"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF09",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: Must cover at least one experimental finding. Verification method: Locate the experimental evidence page and check that at least one finding or qualitative trend is presented. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 5: at least one experimental finding.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G05"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF10",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: Must cover the boundaries of the conclusions. Verification method: Locate the conclusion or boundary page and check whether it states the conditions under which the conclusions apply. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 5: the boundaries of the conclusions.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G05"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF11",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: Must preserve the theorem's restrictions on scale/models. Verification method: Search page by page for whether the scale/model-related qualifications are preserved where the theory is stated. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 7: Preserve theorem restrictions on scale/models.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G06"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF12",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: Must preserve the conclusions' dependencies on conjectures. Verification method: Check whether the relevant theory/depth discussion notes the dependencies on conjectures. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 7: dependencies on conjectures.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G06"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF13",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: Must distinguish expressivity theory from training experiments. Verification method: Check that expressivity theory and training experiments are distinguishable in wording, qualifications, or grouping; they may be placed side by side on the same page, and splitting them onto different pages is not required; training trends cannot serve as theorem proofs. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 7: Distinguish expressivity theory from training experiments.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G06"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF14",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: Must not generalize the conclusions into a performance ranking across all real-world tasks. Verification method: Search for whether any all-task performance ranking or universal superiority/inferiority assertion appears. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 7: do not generalize to a performance ranking across all real-world tasks.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G06"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF15",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: Must reorganize the poster into a narrative that can be delivered orally, rather than cutting it into six full-slide poster screenshots. Verification method: Check whether the six pages have a reorganized presentation order and explanatory text, and confirm that they are not six full-slide poster screenshots. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 5: Reorganize it into a spoken narrative rather than cutting it into six full-slide screenshots..",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G05"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF16",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: Key explanations must remain editable text and graphics. Verification method: In the PPTX, click on the key explanatory text and diagrams and confirm that they are text boxes/shapes/charts rather than only images. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 9: key explanations must remain editable text and graphics.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G07"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF17",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: Must not rasterize entire slides. Verification method: Check whether any page has a single image covering the entire page, and confirm that no entire slide is rasterized. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 9: do not rasterize entire slides.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G08"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF18",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: The original input files in input/ must remain read-only and must not be modified. [v7 verifiable boundary: Check that the final inputs match the frozen snapshot file by file; proving that no writes ever occurred during production is not required. Inputs at public runtime are guaranteed to be read-only by the read-only mount.]. Verification method: Check the obtainable evidence of input integrity against the content/hashes of the frozen originals; the read-only mount at public runtime can serve as a guarantee. File timestamps, missing history logs, or unpreserved intermediate processes must not be used to infer rewriting. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 11: Original input files are read-only..",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G09"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF19",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: Must deliver six slide PNGs. Verification method: Check that six PNGs exist in the delivery area and correspond to the six PPTX pages. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 11: six slide PNGs.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G10"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF20",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: Must deliver sources.md. Verification method: Check that the sources.md file exists and is readable. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 11: sources.md (sources, versions, and figure provenance).",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G11"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF21",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: sources.md must list the sources of the facts and figures actually used, including the author's poster and the corresponding paper. Verification method: Check that the author's poster and the corresponding paper in sources.md can be located and correspond to the facts or figures actually used; do not judge No merely because a reading-scope explanation file is not listed as a source. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 11: sources.md (sources, versions, and figure provenance).",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G11"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF22",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: sources.md must record source versions. Verification method: Check that sources.md lists version information, such as paper arXiv v1, poster ICML 2024. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 11: sources.md (sources, versions, and figure provenance).",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G11"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF23",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: sources.md must record figure provenance. Verification method: Check that sources.md records the provenance of each poster/paper figure used in the PPTX. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 11: sources.md (sources, versions, and figure provenance).",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G11"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF24",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: sources.md must identify the version differences between the poster and the paper. Verification method: Check that sources.md points out the version differences between the poster and the paper (e.g., ICML 2024 poster vs arXiv v1 paper). Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Material basis (relative to inputs/): input/reading-scope.md original line 5: Identify version differences.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G11"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF25",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: Must not attribute content from a later poster version to proofs already present in arXiv v1. Verification method: Check the statements in the PPTX and sources.md that involve later poster content, and confirm that they do not claim this content was already proved by v1. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Material basis (relative to inputs/): input/reading-scope.md original line 5: do not attribute later poster content to proofs already in v1.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G12"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF26",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: Must deliver delivery.md. Verification method: Check that the delivery.md file exists and is readable. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Task basis: prompt.md original line 11: and delivery.md.",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G11"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "IF28",
    "dimension": "Task Fulfillment / Compliance",
    "criterion": "Requirement: The PPTX must adapt the main content of the author's conference poster input/poster.png. Verification method: Compare the PPTX topics with the main content of input/poster.png, and confirm that the research question, connection, depth/task, experiments, limitations, and other content come from adapting the poster. Pass criterion: Fully satisfies this requirement; missing, incorrect, or partial satisfaction is No. A failure of the evaluation tool cannot be the basis for a No. Verification boundary: Verify against the frozen task statement and the necessary originals/ground truth; the author's self-description does not replace artifact evidence. Material basis (relative to inputs/): input/reading-scope.md original line 3: Adapt the main content of the author's conference poster, poster.png..",
    "answers": {
      "Yes": "Complete all necessary checks for this criterion, with actual evidence supporting satisfaction of the requirement; equivalent implementations within the task's scope are allowed.",
      "No": "At least one necessary requirement is not satisfied; identify evidence of actual omissions, errors, or partial satisfaction."
    },
    "compliance_groups": [
      "G13"
    ],
    "benchmark_source": "vas"
  },
  {
    "id": "C1",
    "dimension": "Content",
    "criterion": "Accurate expression of terminology and theoretical conditions.\nObject of measurement: Terms such as MPC, parallel computation, k-hop, pointer doubling, depth upper/lower bounds, and near-equivalence must be used accurately and in a way suitable for an audience familiar with Transformer fundamentals; theoretical conditions (resources, models, conjectures) must not be misused by terminology or swallowed in the presentation.\nObservation scope: The 6-page English PPTX and PNGs, adapting the author's poster into an oral presentation for a research group.\nObservation basis: Extract terms such as MPC, near-equivalence, k-hop, pointer doubling, and upper/lower bounds page by page, and check their usage against paper §1.1 and §3—5. Check whether statements of sufficiency and conditional lower bounds are stated separately, and whether phrasing such as \"unconditional equivalence\" appears.\nScoring boundary: Whether terminology, concepts, and phrasing allow this task's audience to understand accurately; IF checks factual realization, and C1 does not award points based on the number of terms.\nOriginal full refinement inspection basis: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; the most easily confused concepts and qualifications remain concise and accurate, and there is no meaning drift among summary, detail, and actual examples.\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.",
    "answers": {
      "-3": "Severe failure: The core understanding or usage goal of this item cannot be achieved. Specific manifestation: Core terminology is misused or replaced with unrelated concepts, and the audience cannot identify the Transformer–MPC and k-hop discussion. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "-2": "Clearly unacceptable: The main area or key process of this item is continuously impaired. Specific manifestation: Multiple key terminology errors, or sufficiency and conditional lower bounds are conflated. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "-1": "Below basic requirements: This item has a clear local substantial defect. Specific manifestation: Individual terms are not accurate enough or are not explained, affecting one key understanding. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "0": "Basically acceptable: Terminology is correct, key concepts are explained, and the audience can follow; the necessary requirements of this item are fully satisfied with no substantial defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: terminology is used consistently and in a targeted way, and key concepts are accurately anchored in one sentence. Quality improvement must be reflected as: easily confused terms, quantities, or references are briefly clarified where they appear, reducing the need to guess. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: the terminology system connects well with the audience's prior knowledge, and theoretical conditions are accurately qualified at the point of definition. Quality improvement must be reflected as: definitions, qualifications, and actual labels support one another, so complex meanings require no cross-region assembly. terminology, symbols, and conditional statements are mature and unified, concise without sacrificing accuracy, reaching a level suitable for external presentation. The most easily confused concepts and qualifications remain concise and accurate, and there is no meaning drift among summary, detail, and actual examples. Verify within the complex parts relevant to this item: the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required. Task-specific object: Terms such as MPC, parallel computation, k-hop, pointer doubling, depth upper/lower bounds, and near-equivalence must be used accurately and in a way suitable for an audience familiar with Transformer fundamentals; theoretical conditions (resources, models, conjectures) must not be misused by terminology or swallowed in the presentation. Difficult-case and inspection basis: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; the most easily confused concepts and qualifications remain concise and accurate, and there is no meaning drift among summary, detail, and actual examples. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. An unusually economical conceptual formulation makes a genuinely confusable distinction directly understandable, with less interpretation than an equally accurate, concise conventional explanation. Identify the distinction and the wording/representation that produces this extra understanding; merely correct terminology or fluent prose is not exceptional. Apply the exceptional benefit specifically to: Terms such as MPC, parallel computation, k-hop, pointer doubling, depth upper/lower bounds, and near-equivalence must be used accurately and in a way suitable for an audience familiar with Transformer fundamentals; theoretical conditions (resources, models, conjectures) must not be misused by terminology or swallowed in the presentation. In-task verification cases for that benefit: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; the most easily confused concepts and qualifications remain concise and accurate, and there is no meaning drift among summary, detail, and actual examples. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 9,
    "benchmark_source": "vas"
  },
  {
    "id": "C2",
    "dimension": "Content",
    "criterion": "Correspondence between claims and paper/poster evidence.\nObject of measurement: Claims about the research question, the Transformer–MPC simulation relationship, the k-hop and depth relationship, and experimental findings must be consistent with what paper v1 and the poster can support; when experimental trends are stated qualitatively, they must not be presented as theoretical proofs or all-task performance conclusions.\nObservation scope: The 6-page English PPTX and PNGs, adapting the author's poster into an oral presentation for a research group.\nObservation basis: Compare each page's factual statements against paper §1.1, §3—5, and the corresponding poster modules, marking supported/overreaching/unsupported. Check whether the experiment page presents the N=100 error trend for L=2 to 6 as a general performance ranking or theoretical proof.\nScoring boundary: The support relationship between task claims and evidence/data/actual process; do not repeatedly evaluate graphical encoding or whether source documents are easy to find.\nOriginal full refinement inspection basis: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; the most complex key comparisons or state chains can still be checked item by item; approximations, limitations, and counterexamples are not concealed by the conclusion.\nScenario reference · content quality (adds no threshold): Research question; Transformer–MPC connection and its resource limitations; k-hop task and depth relationship (including conjecture dependencies); at least one experimental finding; boundaries of the conclusions.\nApplicable boundary for this task (applies to all relevant items; adds no scoring items): Does not require incorporating \"Subsequent improvement\" into the conclusions of paper v1.\nApplicable boundary for this task (applies to all relevant items; adds no scoring items): Preserve the distinction between \"expressivity theory\" and \"training experiments\", and do not generalize from it to a real-task performance ranking.\nApplicable boundary for this task (applies to all relevant items; adds no scoring items): \"Subsequent improvement\" is follow-up work and must not be attributed to the proofs in the frozen paper v1.\nApplicable boundary for this task (applies to all relevant items; adds no scoring items): At least one piece of experimental evidence (the upper-right error plot or the middle-right attention structure) must be presented; both are not required; a qualitative trend suffices.\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.",
    "answers": {
      "-3": "Severe failure: The core understanding or usage goal of this item cannot be achieved. Specific manifestation: Core claims are opposite to the paper or unsupported by sources, and experiments/theory are conflated to the point of changing the conclusion. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "-2": "Clearly unacceptable: The main area or key process of this item is continuously impaired. Specific manifestation: Multiple claims overreach, such as treating training experiments as unconditional proof, or generalizing limited conclusions into failure of all architectures. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "-1": "Below basic requirements: This item has a clear local substantial defect. Specific manifestation: One claim lacks qualification, or later work from the poster is mistakenly attributed to paper v1. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "0": "Basically acceptable: Necessary claims are supported and boundaries are correct; the necessary requirements of this item are fully satisfied with no substantial defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: key claims can all be mapped to specific charts or sections, and qualifiers are used appropriately. Quality improvement must be reflected as: key claims correspond closely to the numerical values, conditions, or process evidence used, making verification more direct. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: theoretical and experimental evidence are cited in layers, and qualitative trends and conditional conclusions support each other without overreaching. Quality improvement must be reflected as: evidence selection, comparison scope, and conclusion strength work together, making clear what can and cannot be inferred. all verifiable claims have precise evidence anchors, and versions and uncertainty are handled transparently, reaching review-level soundness. The most complex key comparisons or state chains can still be checked item by item; approximations, limitations, and counterexamples are not concealed by the conclusion. Verify within the complex parts relevant to this item: the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required. Task-specific object: Claims about the research question, the Transformer–MPC simulation relationship, the k-hop and depth relationship, and experimental findings must be consistent with what paper v1 and the poster can support; when experimental trends are stated qualitatively, they must not be presented as theoretical proofs or all-task performance conclusions. Difficult-case and inspection basis: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; the most complex key comparisons or state chains can still be checked item by item; approximations, limitations, and counterexamples are not concealed by the conclusion. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. A non-obvious presentation of the evidence exposes an important conditional relationship that a flawless fact-and-citation account would leave the reader to reconstruct. Show the additional understanding and its support, without inventing a claim, requiring more data, or rewarding the count of sources. Apply the exceptional benefit specifically to: Claims about the research question, the Transformer–MPC simulation relationship, the k-hop and depth relationship, and experimental findings must be consistent with what paper v1 and the poster can support; when experimental trends are stated qualitatively, they must not be presented as theoretical proofs or all-task performance conclusions. In-task verification cases for that benefit: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; the most complex key comparisons or state chains can still be checked item by item; approximations, limitations, and counterexamples are not concealed by the conclusion. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 9,
    "benchmark_source": "vas"
  },
  {
    "id": "C3",
    "dimension": "Content",
    "criterion": "Information selection and tone for an oral presentation.\nObject of measurement: Select poster information according to the purpose of a six-page oral presentation: cover the five content items required by the task statement, preserve necessary qualifications, and delete or downgrade non-essential details; the tone should be academic narration rather than poster listing.\nObservation scope: The 6-page English PPTX and PNGs, adapting the author's poster into an oral presentation for a research group.\nObservation basis: Check page by page whether the research question, Transformer–MPC, k-hop and depth, at least one experiment, and the boundaries of the conclusions all appear. Check whether there is wholesale copying of poster text or deletion of conjectures/resource conditions for compression.\nScoring boundary: Content selection, density, tone, and necessary qualifications around the scenario purpose; not font size or page visual hierarchy.\nOriginal full refinement inspection basis: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; the densest information still has appropriate selection, and key points and exceptions are understandable, with no redundant expansion or oversimplification.\nTask scenario: An internal academic presentation in a research group familiar with Transformer fundamentals: the author adapts the ICML 2024 conference poster and the arXiv v1 paper into an English six-page PPTX, orally presenting the research question, theoretical connection, synthetic tasks, experimental evidence, and boundaries of the conclusions, and delivers PNGs, sources, and delivery instructions.\nScenario reference · audience (adds no threshold): Research group listeners familiar with basic Transformer mechanisms but not necessarily with MPC and parallel complexity theory.\nScenario reference · purpose (adds no threshold): An approximately six-page oral presentation that organizes evidence with a spoken narrative rather than a poster screenshot carousel.\nScenario reference · content selection (adds no threshold): The argument hierarchy is clear, formulas and charts are readable, and evidence precedes decoration; qualitative experimental trends are sufficient, and theoretical conditions and conjecture dependencies are indispensable.\nApplicable boundary for this task (applies to all relevant items; adds no scoring items): Does not require reproducing the threshold symbols in the upper-right Result of the poster, nor reproducing all theorem parameters or exact threshold formulas.\nApplicable boundary for this task (applies to all relevant items; adds no scoring items): Does not require presenting both the upper-right error plot and the middle-right attention structure; either one suffices.\nApplicable boundary for this task (applies to all relevant items; adds no scoring items): Does not require putting all facts on the slides, nor treat the language of the Chinese reference artifact as a requirement.\nApplicable boundary for this task (applies to all relevant items; adds no scoring items): Preserve the resource-condition qualification of near-equivalence; it must not be written as unconditional complete equivalence.\nContent focus: Transformer–MPC, k-hop depth relationship, expressivity theory and training experiments, versions and conjecture qualifications\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.",
    "answers": {
      "-3": "Severe failure: The core understanding or usage goal of this item cannot be achieved. Specific manifestation: Multiple items among the research question/theoretical connection/k-hop depth/experiments/boundaries are not covered, or the full poster text is compressed into dense text. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "-2": "Clearly unacceptable: The main area or key process of this item is continuously impaired. Specific manifestation: Two or more required content items are missing, or key qualifications are deleted, causing the tone of the conclusion to be distorted. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "-1": "Below basic requirements: This item has a clear local substantial defect. Specific manifestation: One required content item is mentioned only in passing, or the tone still resembles a poster list. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "0": "Basically acceptable: All five content items are covered, with reasonable detail; the necessary requirements of this item are fully satisfied with no substantial defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: selection shows a clear awareness of oral narrative, and non-key details give way to the argument. Quality improvement must be reflected as: removing restatements that do not aid understanding or placing necessary qualifications where they are needed, making the key points more direct. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: the information volume of each page is designed according to the presentation rhythm, and qualifications and evidence complement each other in detail. Quality improvement must be reflected as: the hierarchy, depth of explanation, and tone suit the audience; the streamlined content still retains necessary conditions. selection precisely serves the oral argument, and both deletion and retention can be explained from the audience's understanding, showing mature restraint. The densest information still has appropriate selection, and key points and exceptions are understandable, with no redundant expansion or oversimplification. Verify within the complex parts relevant to this item: the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required. Task-specific object: Select poster information according to the purpose of a six-page oral presentation: cover the five content items required by the task statement, preserve necessary qualifications, and delete or downgrade non-essential details; the tone should be academic narration rather than poster listing. Difficult-case and inspection basis: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; the densest information still has appropriate selection, and key points and exceptions are understandable, with no redundant expansion or oversimplification. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. An unusually incisive selection resolves the tension between necessary qualifications and the audience’s main purpose: it conveys more useful understanding with no loss of required substance. Explain why a conventionally fully refined summary would still impose more interpretation; brevity, completeness or promotional impact alone is insufficient. Apply the exceptional benefit specifically to: Select poster information according to the purpose of a six-page oral presentation: cover the five content items required by the task statement, preserve necessary qualifications, and delete or downgrade non-essential details; the tone should be academic narration rather than poster listing. In-task verification cases for that benefit: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; the densest information still has appropriate selection, and key points and exceptions are understandable, with no redundant expansion or oversimplification. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 7,
    "benchmark_source": "vas"
  },
  {
    "id": "O1",
    "dimension": "Organization",
    "criterion": "Six-page argument path.\nObject of measurement: The six pages should form a coherent oral argument path from the research question to the theoretical connection, tasks and depth, experimental evidence, and boundaries, and the page order should let the audience anticipate and follow the progression of the argument.\nObservation scope: The 6-page English PPTX and PNGs, adapting the author's poster into an oral presentation for a research group.\nObservation basis: Record each page's claims page by page, draw the dependency relationships among claims, and check whether any theory or experiment appears without preparation. Check whether the cover/conclusion serve the argument rather than being merely decorative.\nScoring boundary: The logic of the overall argument/decision/operation/concept path, not scored against a fixed structural template.\nOriginal full refinement inspection basis: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; the most complex branches or comparisons can still be incorporated into the main line, and the closure is grounded, with no dangling steps or repeated detours.\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.",
    "answers": {
      "-3": "Severe failure: The core understanding or usage goal of this item cannot be achieved. Specific manifestation: The page order is illogical, and the audience cannot reconstruct the main line of the argument. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "-2": "Clearly unacceptable: The main area or key process of this item is continuously impaired. Specific manifestation: Multiple jumps or repetitions, and the order of theory and experiment pages creates obstacles to understanding. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "-1": "Below basic requirements: This item has a clear local substantial defect. Specific manifestation: One transition is abrupt, or one page carries too many unrelated points. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "0": "Basically acceptable: The page order is reasonable, and the opening, development, turn, and conclusion are identifiable; the necessary requirements of this item are fully satisfied with no substantial defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: the argument path is clear, and transitions between pages have a clear direction. Quality improvement must be reflected as: one key turn explicitly connects the premise to the next step, making the main line easier to follow. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: each page has a clear position in the overall argument, and the progression of theory and experiments prepares for each other. Quality improvement must be reflected as: the premises, development, and conclusions of each part refer back to one another, and the order and length serve the explanation or decision of this task. the six-page path is compact and has a presentation rhythm, and omitting pages does not harm the main line, showing mature academic narrative. The most complex branches or comparisons can still be incorporated into the main line, and the closure is grounded, with no dangling steps or repeated detours. Verify within the complex parts relevant to this item: the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required. Task-specific object: The six pages should form a coherent oral argument path from the research question to the theoretical connection, tasks and depth, experimental evidence, and boundaries, and the page order should let the audience anticipate and follow the progression of the argument. Difficult-case and inspection basis: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; the most complex branches or comparisons can still be incorporated into the main line, and the closure is grounded, with no dangling steps or repeated detours. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. A non-obvious organization of the existing argument or user path makes separate necessary parts illuminate one another and materially reduces reasoning or navigation steps. Compare the benefit with an equally complete, logically sound conventional order; no additional section or prescribed order is required. Apply the exceptional benefit specifically to: The six pages should form a coherent oral argument path from the research question to the theoretical connection, tasks and depth, experimental evidence, and boundaries, and the page order should let the audience anticipate and follow the progression of the argument. In-task verification cases for that benefit: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; the most complex branches or comparisons can still be incorporated into the main line, and the closure is grounded, with no dangling steps or repeated detours. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 6,
    "benchmark_source": "vas"
  },
  {
    "id": "O2",
    "dimension": "Organization",
    "criterion": "Local semantic development within a page.\nObject of measurement: The relationship among the points, evidence, and explanation within each page must be clear: the title, key sentences, figures/formulas, and annotations form a readable local argument rather than a stack of parallel information.\nObservation scope: The 6-page English PPTX and PNGs, adapting the author's poster into an oral presentation for a research group.\nObservation basis: Check whether each page has a clear claim sentence, and whether figures/formulas have explanatory text rather than only titles. Check whether local annotations explain the relationship between evidence and claims.\nScoring boundary: The semantic order and relationships of explanation within a page, stage, node, or comparison unit.\nOriginal full refinement inspection basis: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; the most complex units can also be understood in the required order, and local explanations are concise, self-contained, and consistent with the overall main line.\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.",
    "answers": {
      "-3": "Severe failure: The core understanding or usage goal of this item cannot be achieved. Specific manifestation: Information within the page is unrelated, and the reader cannot determine which sentence the evidence supports. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "-2": "Clearly unacceptable: The main area or key process of this item is continuously impaired. Specific manifestation: Multiple places have only titles and materials side by side, lacking explanatory relationships. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "-1": "Below basic requirements: This item has a clear local substantial defect. Specific manifestation: The relationship between evidence and claims within one page requires guessing. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "0": "Basically acceptable: Each page shows the necessary points and supporting materials; the necessary requirements of this item are fully satisfied with no substantial defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: local development has layers, and figure captions/short sentences clearly point to the points. Quality improvement must be reflected as: the arrangement of objects, explanations, and results within a local unit reduces one necessary cross-check. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: points—evidence—qualifications form a compact chain within the page, and formulas or charts are explained in place. Quality improvement must be reflected as: the figures, text, conditions, and conclusions of each major unit form relationships understandable in place. the semantic organization within the page is refined and complete, requiring almost no additional patching during oral presentation. The most complex units can also be understood in the required order, and local explanations are concise, self-contained, and consistent with the overall main line. Verify within the complex parts relevant to this item: the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required. Task-specific object: The relationship among the points, evidence, and explanation within each page must be clear: the title, key sentences, figures/formulas, and annotations form a readable local argument rather than a stack of parallel information. Difficult-case and inspection basis: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; the most complex units can also be understood in the required order, and local explanations are concise, self-contained, and consistent with the overall main line. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. An economical local organization makes a difficult relationship among conditions, evidence and objects understandable as one unit rather than assembled mentally from carefully grouped parts. Identify the semantic reconstruction it eliminates; proximity and tidy grouping alone do not qualify. Apply the exceptional benefit specifically to: The relationship among the points, evidence, and explanation within each page must be clear: the title, key sentences, figures/formulas, and annotations form a readable local argument rather than a stack of parallel information. In-task verification cases for that benefit: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; the most complex units can also be understood in the required order, and local explanations are concise, self-contained, and consistent with the overall main line. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 5,
    "benchmark_source": "vas"
  },
  {
    "id": "O3",
    "dimension": "Organization",
    "criterion": "Inheritance of terminology, symbols, and conclusions across pages.\nObject of measurement: The same concept, symbol, and qualification must remain consistent across the six pages and be inherited page by page: k-hop, depth, upper/lower bounds, near-equivalence, etc. must not be renamed or lose qualifications; the conclusion page must recall earlier evidence rather than introduce new unspoken claims.\nObservation scope: The 6-page English PPTX and PNGs, adapting the author's poster into an oral presentation for a research group.\nObservation basis: Build a terminology/symbol comparison table and check the usage of k-hop, L, N, upper/lower bounds, and near-equivalence page by page. Check whether each conclusion on the conclusion page has corresponding evidence in the first five pages.\nScoring boundary: Semantic inheritance across pages/states, or the closure of a single page's branch path with identity/action; forcing a new ending is not required.\nOriginal full refinement inspection basis: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; complex branches, final states, or resets still maintain identity and causal continuity, and the reader does not need later information to repair the current meaning.\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.\nEvaluate meaning carried across the actual slides, not invented software branches or reset states.",
    "answers": {
      "-3": "Severe failure: The core understanding or usage goal of this item cannot be achieved. Specific manifestation: The same core concept is contradictory before and after, or the conclusion page presents new claims unsupported earlier. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "-2": "Clearly unacceptable: The main area or key process of this item is continuously impaired. Specific manifestation: Multiple terminology/symbol inconsistencies, making cross-page comparison difficult. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "-1": "Below basic requirements: This item has a clear local substantial defect. Specific manifestation: One key qualification is lost on a later page or a symbol's meaning changes. Judge according to the actual impact above; harmless minor differences do not trigger this tier.",
      "0": "Basically acceptable: Terminology is consistent across pages, and conclusions correspond to earlier text; the necessary requirements of this item are fully satisfied with no substantial defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: symbols and qualifications have stable inheritance, and the conclusion page clearly recalls evidence. Quality improvement must be reflected as: one cross-page, cross-region, or cross-state handoff clearly preserves object identity and existing meaning. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: cross-page identity and semantic inheritance are clear, and the theory and experiment lines converge correctly on the conclusion page. Quality improvement must be reflected as: the main handoffs, branches, and ending form continuous semantics; the positions of repeated objects and necessary exceptions are traceable. cross-page inheritance is rigorous and concise, and qualifications are not lost in compressed statements, showing mature consistency. Complex branches, final states, or resets still maintain identity and causal continuity, and the reader does not need later information to repair the current meaning. Verify within the complex parts relevant to this item: the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required. Task-specific object: The same concept, symbol, and qualification must remain consistent across the six pages and be inherited page by page: k-hop, depth, upper/lower bounds, near-equivalence, etc. must not be renamed or lose qualifications; the conclusion page must recall earlier evidence rather than introduce new unspoken claims. Difficult-case and inspection basis: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; complex branches, final states, or resets still maintain identity and causal continuity, and the reader does not need later information to repair the current meaning. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. A purposeful carry-over lets the audience preserve and develop the right meaning through the task’s actual transitions, materially reducing reconstruction beyond fully refined referential consistency. Identify what understanding accumulates; do not invent branches, endings or reset states absent from the task. Apply the exceptional benefit specifically to: The same concept, symbol, and qualification must remain consistent across the six pages and be inherited page by page: k-hop, depth, upper/lower bounds, near-equivalence, etc. must not be renamed or lose qualifications; the conclusion page must recall earlier evidence rather than introduce new unspoken claims. In-task verification cases for that benefit: Check only within this item's applicable scope the pages densest in method/theorem conditions and the recall of experimental conclusions; listing all theorem parameters is not required; complex branches, final states, or resets still maintain identity and causal continuity, and the reader does not need later information to repair the current meaning. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 4,
    "benchmark_source": "vas"
  },
  {
    "id": "V1",
    "dimension": "Visuals",
    "criterion": "On-page visual hierarchy and focus.\nWhat is measured: Each page should let the audience first see that page's core claim or key evidence; titles, claim sentences, figures/formulas, and supporting information form a clear hierarchy; decoration does not steal the focus from evidence.\nScope of observation: 6-page English PPTX and PNG, adapting the author's poster into an oral presentation for a research group.\nBasis for observation: Using thumbnails and 100% view separately, observe whether each page's first landing point is the core claim/evidence. Check whether the area and saturation of decorative elements overpower charts or formulas.\nScoring boundary: The salience relationship among title, body, evidence, and supporting information; evaluate the attention entry point, core focus, and emphasis strength.\nHigh-band verification: Check only within this item's applicable scope the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters; even in the densest region or state, the main information is seen before the basis and qualifications, and there is no substantive attention competition between the whole and the parts.\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.\nJudge negative visual grades by substantive harm and extent within this criterion, not a prerequisite of functional unusability. Harmless style differences are not defects; retain the original concrete defect examples.",
    "answers": {
      "-3": "Severe failure: This item's core understanding or use goal cannot be completed. Specific manifestation: no visible focus; title, body text, and figures compete with one another to the point of being unreadable. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-2": "Clearly unacceptable: This item's main area or key process is persistently impaired. Specific manifestation: focus lands on decoration or secondary information, and core evidence is drowned out. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-1": "Below basic requirements: This item has a clear local substantive defect. Specific manifestation: hierarchy is weak, and the audience must expend effort to find the page's key point. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "0": "Meets basic requirements: Each page has a recognizable title and necessary content area, and focus holds; this item's necessary requirements are fully met, with no substantive defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: Meets band 0; the claim sentence or key figure has clear emphasis, and supporting information recedes. Quality improvement must be manifested as: a set of position, scale, or contrast choices that make the primary focus clearer, with supporting information not competing for reading attention. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: Meets band +1; hierarchy changes according to argumentative needs, and formula/chart and textual focus complement each other. Quality improvement must be manifested as: the strong/weak relationships among primary focus, evidence, and necessary qualifications are coordinated, and adjusted reasonably with content complexity. Meets band +2; the hierarchy design is mature, the viewing path is natural when presenting, evidence always dominates and is not rigid. Even in the densest region or state, the main information is seen before the basis and qualifications, and there is no substantive attention competition between the whole and the parts. Verify in this item's relevant complex parts: the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters. Task-specific object: Each page should let the audience first see that page's core claim or key evidence; titles, claim sentences, figures/formulas, and supporting information form a clear hierarchy; decoration does not steal the focus from evidence. Difficult-case and inspection basis: Using thumbnails and 100% view separately, observe whether each page's first landing point is the core claim/evidence. Check whether the area and saturation of decorative elements overpower charts or formulas. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. A non-obvious hierarchy brings the decisive relationship and its necessary qualification into attention together without competing emphasis. Demonstrate a first-look interpretation benefit beyond an impeccably clear primary/secondary hierarchy, not just stronger contrast or larger type. Apply the exceptional benefit specifically to: Each page should let the audience first see that page's core claim or key evidence; titles, claim sentences, figures/formulas, and supporting information form a clear hierarchy; decoration does not steal the focus from evidence. In-task verification cases for that benefit: Using thumbnails and 100% view separately, observe whether each page's first landing point is the core claim/evidence. Check whether the area and saturation of decorative elements overpower charts or formulas. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 5,
    "benchmark_source": "vas"
  },
  {
    "id": "V2",
    "dimension": "Visuals",
    "criterion": "Block proportions, alignment, and whitespace.\nWhat is measured: Page block division, text/figure proportions, margins, and alignment should suit projected reading during an oral presentation; whitespace serves grouping, rather than being uniformly filled or randomly vacant.\nScope of observation: 6-page English PPTX and PNG, adapting the author's poster into an oral presentation for a research group.\nBasis for observation: Measure or visually estimate whether page margins, title area, body area, and annotation area are aligned consistently. Check whether charts and text boxes have unreasonable crowding or large areas of meaningless blank space.\nScoring boundary: Area proportions among information blocks, grid, edges, visual center of gravity, density, and whitespace; evaluate overall composition and spatial organization.\nHigh-band verification: Check only within this item's applicable scope the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters; regions with the greatest density or state variation still have stable spatial order, and overall balance and local spacing both hold.\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.\nJudge negative visual grades by substantive harm and extent within this criterion, not a prerequisite of functional unusability. Harmless style differences are not defects; retain the original concrete defect examples.",
    "answers": {
      "-3": "Severe failure: This item's core understanding or use goal cannot be completed. Specific manifestation: blocks are chaotic, severely overlapping, or cropped, and the page cannot be read normally. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-2": "Clearly unacceptable: This item's main area or key process is persistently impaired. Specific manifestation: multiple instances of imbalanced text/figure proportions or broken alignment interfere with reading evidence. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-1": "Below basic requirements: This item has a clear local substantive defect. Specific manifestation: whitespace or alignment on several pages is clearly arbitrary, increasing reading burden. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "0": "Meets basic requirements: Page zones are discernible, and margins and alignment are normal; this item's necessary requirements are fully met, with no substantive defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: Meets band 0; block proportions are stable, and whitespace effectively separates claims from evidence. Quality improvement must be manifested as: block edges, scale, or whitespace have a clear purpose, improving the spatial relationship of one class of content. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: Meets band +1; different page types have a unified yet flexible grid, and text/figure proportions adjust appropriately with content. Quality improvement must be manifested as: area proportions, alignment, visual center of gravity, and whitespace jointly adapt to actual content, rather than mechanically applying the same layout. Meets band +2; layout rhythm is mature, and whitespace and block proportions precisely support the presentation, with almost nothing to criticize. Regions with the greatest density or state variation still have stable spatial order, and overall balance and local spacing both hold. Verify in this item's relevant complex parts: the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters. Task-specific object: Page block division, text/figure proportions, margins, and alignment should suit projected reading during an oral presentation; whitespace serves grouping, rather than being uniformly filled or randomly vacant. Difficult-case and inspection basis: Measure or visually estimate whether page margins, title area, body area, and annotation area are aligned consistently. Check whether charts and text boxes have unreasonable crowding or large areas of meaningless blank space. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. A content-specific spatial solution reconciles competing demands of overview, detail and available space with exceptional clarity and balance. Identify the real space conflict resolved beyond a flawlessly fitted conventional composition; extra whitespace, density or a fashionable layout is not itself a benefit. Apply the exceptional benefit specifically to: Page block division, text/figure proportions, margins, and alignment should suit projected reading during an oral presentation; whitespace serves grouping, rather than being uniformly filled or randomly vacant. In-task verification cases for that benefit: Measure or visually estimate whether page margins, title area, body area, and annotation area are aligned consistently. Check whether charts and text boxes have unreasonable crowding or large areas of meaningless blank space. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 8,
    "benchmark_source": "vas"
  },
  {
    "id": "V3",
    "dimension": "Visuals",
    "criterion": "Fonts, type sizes, and formula/chart readability.\nWhat is measured: The type sizes and line spacing of titles, body text, figure captions, and formulas are readable at projection scale; line breaks do not break terms or formulas; key formulas and chart labels are legible at normal viewing distance.\nScope of observation: 6-page English PPTX and PNG, adapting the author's poster into an oral presentation for a research group.\nBasis for observation: Check each page's minimum font size and formula labels at projection resolution/normal viewing distance. Check whether terms, variable names, and formulas are improperly broken across lines or compressed/distorted.\nScoring boundary: The complete visibility of text in the actually delivered image, type size, contrast, line spacing, and line breaks; evaluate text presentation and reading at the target scale.\nHigh-band verification: Check only within this item's applicable scope the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters; the longest necessary labels, small qualifications, or complex symbols remain complete and easy to read, and typographic quality remains consistently stable across all regions.\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.\nJudge negative visual grades by substantive harm and extent within this criterion, not a prerequisite of functional unusability. Harmless style differences are not defects; retain the original concrete defect examples.",
    "answers": {
      "-3": "Severe failure: This item's core understanding or use goal cannot be completed. Specific manifestation: body text or key formulas are too small to read, and term line breaks cause misunderstanding. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-2": "Clearly unacceptable: This item's main area or key process is persistently impaired. Specific manifestation: multiple instances of inappropriate type size/line spacing; formulas or figure captions require zooming to read. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-1": "Below basic requirements: This item has a clear local substantive defect. Specific manifestation: individual pages have slightly small type or poor line breaks, affecting local reading. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "0": "Meets basic requirements: Titles and body text are readable, and key formulas/figure captions are legible; this item's necessary requirements are fully met, with no substantive defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: Meets band 0; the type-size hierarchy is clear, and line spacing and line-break handling are stable. Quality improvement must be manifested as: the type size, spacing, or line breaks of one class of necessary text reduce reading burden; being readable alone is not a sufficient basis. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: Meets band +1; at projection scale, formulas and chart labels are comfortably readable, and line breaks respect terms. Quality improvement must be manifested as: type size, line spacing, contrast, and semantic line breaks are coordinated, and dense content remains comfortably readable at the scale specified by the prompt. Meets band +2; the typographic scale is mature, formulas and figure captions remain clear amid complex content, and line-spacing rhythm matches the oral presentation. The longest necessary labels, small qualifications, or complex symbols remain complete and easy to read, and typographic quality remains consistently stable across all regions. Verify in this item's relevant complex parts: the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters. Task-specific object: The type sizes and line spacing of titles, body text, figure captions, and formulas are readable at projection scale; line breaks do not break terms or formulas; key formulas and chart labels are legible at normal viewing distance. Difficult-case and inspection basis: Check each page's minimum font size and formula labels at projection resolution/normal viewing distance. Check whether terms, variable names, and formulas are improperly broken across lines or compressed/distorted. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. Typographic structure makes a difficult meaning or relationship apparent as it is read, reducing parsing beyond fully refined legibility, spacing and hierarchy. Identify the specific semantic contribution at the required viewing scale; a distinctive font or decorative lettering is not sufficient. Apply the exceptional benefit specifically to: The type sizes and line spacing of titles, body text, figure captions, and formulas are readable at projection scale; line breaks do not break terms or formulas; key formulas and chart labels are legible at normal viewing distance. In-task verification cases for that benefit: Check each page's minimum font size and formula labels at projection resolution/normal viewing distance. Check whether terms, variable names, and formulas are improperly broken across lines or compressed/distorted. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 7,
    "benchmark_source": "vas"
  },
  {
    "id": "V4",
    "dimension": "Visuals",
    "criterion": "Decoding of figures, arrows, diagrams, and symbols.\nWhat is measured: Poster excerpt figures, k-hop diagrams, depth relationship diagrams, or attention structure diagrams must allow the audience to decode accurately: axes, legends, arrow directions, variable symbols, and figure captions are consistent and do not produce opposite interpretations.\nScope of observation: 6-page English PPTX and PNG, adapting the author's poster into an oral presentation for a research group.\nBasis for observation: Check each figure for consistency among axes, legends, units, variable symbols, and figure captions. Check whether the visual language of theoretical diagrams and experimental figures is mixed in a way that causes misreading.\nScoring boundary: The recognizability and semantic correspondence of graphical marks, lines, arrows, symbols, and dynamic states; evaluate accurate decoding and reliable comparison.\nHigh-band verification: Check only within this item's applicable scope the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters; the most complex connections, states, or scale relationships remain reliably decodable, and there is no ambiguity or misleading between legend and actual content.\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.\nJudge negative visual grades by substantive harm and extent within this criterion, not a prerequisite of functional unusability. Harmless style differences are not defects; retain the original concrete defect examples.",
    "answers": {
      "-3": "Severe failure: This item's core understanding or use goal cannot be completed. Specific manifestation: key encoding errors in charts or misleading arrows/axes lead to the opposite conclusion. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-2": "Clearly unacceptable: This item's main area or key process is persistently impaired. Specific manifestation: multiple missing/inconsistent legends, axes, or symbols make charts difficult to decode. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-1": "Below basic requirements: This item has a clear local substantive defect. Specific manifestation: one figure caption or arrow has an unclear meaning, requiring guessing with verbal supplementation. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "0": "Meets basic requirements: Charts are decodable, and axes/legends/figure captions are complete; this item's necessary requirements are fully met, with no substantive defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: Meets band 0; arrows, variables, and figure captions are consistent, and key figures have in-place explanations. Quality improvement must be manifested as: a set of marks, connections, or contrasts makes object relationships more directly decodable, reducing guesswork about the figure. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: Meets band +1; graphical encoding and textual explanation support each other, and theoretical and experimental diagrams are clearly distinguished. Quality improvement must be manifested as: graphical encoding, annotations, and actual relationships support each other; complex comparisons do not require extra explanation to correct the figure's meaning. Meets band +2; charts and diagrams are decoded precisely and with restraint, complex relationships are readable at a glance, and the result is mature and professional. The most complex connections, states, or scale relationships remain reliably decodable, and there is no ambiguity or misleading between legend and actual content. Verify in this item's relevant complex parts: the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters. Task-specific object: Poster excerpt figures, k-hop diagrams, depth relationship diagrams, or attention structure diagrams must allow the audience to decode accurately: axes, legends, arrow directions, variable symbols, and figure captions are consistent and do not produce opposite interpretations. Difficult-case and inspection basis: Check each figure for consistency among axes, legends, units, variable symbols, and figure captions. Check whether the visual language of theoretical diagrams and experimental figures is mixed in a way that causes misreading. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. A visual encoding makes a difficult relationship directly perceptible instead of requiring mental reconstruction from fully correct labels and marks. Show what becomes easier to understand and verify fidelity to the source or required state; novelty and additional diagrams are not requirements. Apply the exceptional benefit specifically to: Poster excerpt figures, k-hop diagrams, depth relationship diagrams, or attention structure diagrams must allow the audience to decode accurately: axes, legends, arrow directions, variable symbols, and figure captions are consistent and do not produce opposite interpretations. In-task verification cases for that benefit: Check each figure for consistency among axes, legends, units, variable symbols, and figure captions. Check whether the visual language of theoretical diagrams and experimental figures is mixed in a way that causes misreading. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 5,
    "benchmark_source": "vas"
  },
  {
    "id": "V5",
    "dimension": "Visuals",
    "criterion": "Graphical abstraction and typographic integration for academic oral presentations.\nWhat is measured: This item's aesthetics are not only about color scheme: poster excerpts, self-drawn diagrams, formula typography, titles, and headers/footers should form a unified graphical language for an academic presentation; the degree of graphical abstraction, typographic tone, and component style must be integrated, and decoration is subordinate to evidence.\nScope of observation: 6-page English PPTX and PNG, adapting the author's poster into an oral presentation for a research group.\nBasis for observation: Categorize and list each page's visual components: poster bitmaps, self-drawn diagrams, formulas, charts, titles/footers; check whether they share line weights, corner radii, annotation methods, and typographic tone. Check whether the abstraction level of diagrams matches formula/text explanations, and whether decorative icons or flashy graphics overpower evidence. Check whether poster excerpts have unified cropping/annotation treatment rather than being pasted in as-is to create a style discontinuity.\nScoring boundary: Audience, purpose, medium character, and the integration of graphical abstraction, typography, imagery, and component surface language; pure color-scheme aesthetics are delegated to V9.\nHigh-band verification: Check only within this item's applicable scope the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters; even the hardest-to-integrate content harmonizes with the overall visual language; when information is complete, there is still clear and restrained scene expression.\nScenario reference · visual language (adds no new threshold): Academic conference oral presentation character: restrained color scheme, clear typography, unified language of charts and diagrams, decoration subordinate to evidence.\nScenario reference · design constraints (adds no new threshold): The poster's three-column information density conflicts with linear oral narrative; theoretical sufficiency/conditional lower bounds must be kept separate from training experimental evidence; version differences and doubts between the poster and paper v1 must be retained as qualifications.\nScenario reference · design goals (adds no new threshold): The six pages form a tellable argument chain, key terms and conditions are accurate, theoretical and experimental boundaries are distinct, poster material is reorganized into editable explanation rather than cropping, and source doubts are honestly recorded.\nScoring boundary: V9 independently evaluates the aesthetic harmony and refinement of the color scheme; reading impairment caused by text contrast goes to V3, mismatches between color and object meaning go to V4, and cross-page color-code identity drift goes to V7. Cross-item attribution is made only when different functional impairments are demonstrated. V5 cannot merely repeat V9's color-scheme quality assessment, and V9 cannot directly award the professional band merely for \"fitting the scenario.\"\nApplicable boundary of this item (applies to all relevant entries, adds no new scoring item): Do not require replicating the poster's three-column layout or comparing page by page against some reference.\nApplicable boundary of this item (applies to all relevant entries, adds no new scoring item): Do not automatically lower the cap band because of pure color-scheme preference, old style, or whitespace.\nApplicable boundary of this item (applies to all relevant entries, adds no new scoring item): Do not require replicating the poster's three-column layout; the 6-page allocation and visual layout are open.\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.\nJudge negative visual grades by substantive harm and extent within this criterion, not a prerequisite of functional unusability. Harmless style differences are not defects; retain the original concrete defect examples.",
    "answers": {
      "-3": "Severe failure: This item's core understanding or use goal cannot be completed. Specific manifestation: the graphical language is completely fragmented; poster bitmaps, diagram arrows, and text typography look like they come from multiple unrelated documents, and the presentation is interrupted by visual noise. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-2": "Clearly unacceptable: This item's main area or key process is persistently impaired. Specific manifestation: poster screenshots and self-drawn content conflict obviously in style, or extensive decorative graphics overpower evidence, so the whole does not look like a report. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-1": "Below basic requirements: This item has a clear local substantive defect. Specific manifestation: on some pages, graphical abstraction and typographic tone are inconsistent, and poster excerpts look like foreign objects pasted on, requiring local repair. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "0": "Meets basic requirements: The language of poster excerpts, editable text, and diagrams suits an oral presentation for a research group, coexisting harmoniously and not interfering with evidence; this item's necessary requirements are fully met, with no substantive defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: Meets band 0; poster excerpts have undergone cropping, annotation, and border treatment, self-drawn diagrams and formula typography are close in tone, and component styles are basically unified. Quality improvement must be manifested as: specific choices in graphical abstraction, typography, or image treatment improve audience understanding and overall impression. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: Meets band +1; graphical abstraction levels have a clear division of labor: evidence figures, mechanism diagrams, formulas, and text form a complementary visual grammar, and the typographic tone is unified and restrained. Quality improvement must be manifested as: typography, graphics, imagery, or component language coordinates to serve this item's purpose, without substituting nice-looking color schemes for analysis. Meets band +2; the entire report forms a mature and specific academic visual language, with graphical abstraction, image treatment, typography, and component language highly integrated, and decoration always serving evidence and possessing recognizability. Even the hardest-to-integrate content harmonizes with the overall visual language; when information is complete, there is still clear and restrained scene expression. Verify in this item's relevant complex parts: the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters. Task-specific object: This item's aesthetics are not only about color scheme: poster excerpts, self-drawn diagrams, formula typography, titles, and headers/footers should form a unified graphical language for an academic presentation; the degree of graphical abstraction, typographic tone, and component style must be integrated, and decoration is subordinate to evidence. Difficult-case and inspection basis: Categorize and list each page's visual components: poster bitmaps, self-drawn diagrams, formulas, charts, titles/footers; check whether they share line weights, corner radii, annotation methods, and typographic tone. Check whether the abstraction level of diagrams matches formula/text explanations, and whether decorative icons or flashy graphics overpower evidence. Check whether poster excerpts have unified cropping/annotation treatment rather than being pasted in as-is to create a style discontinuity. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. A distinctive but appropriate visual language makes the subject or purpose intrinsic to the graphic form, beyond flawless stylistic integration. Identify a non-color design contribution and its communication benefit; personal taste, modernity and impressive decoration are not evidence. Apply the exceptional benefit specifically to: This item's aesthetics are not only about color scheme: poster excerpts, self-drawn diagrams, formula typography, titles, and headers/footers should form a unified graphical language for an academic presentation; the degree of graphical abstraction, typographic tone, and component style must be integrated, and decoration is subordinate to evidence. In-task verification cases for that benefit: Categorize and list each page's visual components: poster bitmaps, self-drawn diagrams, formulas, charts, titles/footers; check whether they share line weights, corner radii, annotation methods, and typographic tone. Check whether the abstraction level of diagrams matches formula/text explanations, and whether decorative icons or flashy graphics overpower evidence. Check whether poster excerpts have unified cropping/annotation treatment rather than being pasted in as-is to create a style discontinuity. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 7,
    "benchmark_source": "vas"
  },
  {
    "id": "V6",
    "dimension": "Visuals",
    "criterion": "Local geometry, endpoints, and optical alignment.\nWhat is measured: Local craftsmanship such as arrow endpoints, line widths, frame cropping, and icon and text baselines should be clean; diagram connection relationships must not be distorted by misaligned endpoints or cropping.\nScope of observation: 6-page English PPTX and PNG, adapting the author's poster into an oral presentation for a research group.\nBasis for observation: Zoom in to check whether all arrow endpoints land on the target boxes/nodes, and whether line widths are systematic. Check whether image cropping cuts into chart labels or formulas.\nScoring boundary: Graphical outlines, line widths, endpoints, image proportions and cropping, intra-component geometry and optical alignment; evaluate local construction precision.\nHigh-band verification: Check only within this item's applicable scope the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters; the finest connections, complex intersections, or irregular outlines are still handled adequately, and local craftsmanship does not damage the overall effect.\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.\nJudge negative visual grades by substantive harm and extent within this criterion, not a prerequisite of functional unusability. Harmless style differences are not defects; retain the original concrete defect examples.",
    "answers": {
      "-3": "Severe failure: This item's core understanding or use goal cannot be completed. Specific manifestation: key connection lines are broken or wrongly connected, and the geometric relationships in the diagram are distorted. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-2": "Clearly unacceptable: This item's main area or key process is persistently impaired. Specific manifestation: multiple dangling endpoints, cropping that cuts text, or chaotic line widths. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-1": "Below basic requirements: One endpoint, boundary, crop, proportion, or alignment has a clearly visible craftsmanship defect that substantively reduces local construction quality. Provide evidence of actual impact in this item's relevant parts on local geometry, endpoints, and optical alignment; ordinary performance, harmless minor differences, or pure personal preference do not trigger the negative band.",
      "0": "Meets basic requirements: Local geometry is clean, with no misalignment that blocks understanding; this item's necessary requirements are fully met, with no substantive defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: Meets band 0; arrow endpoints, frames, and text alignment withstand close inspection. Quality improvement must be manifested as: the handling of one class of endpoints, line widths, cropping, or optical alignment is more precise than ordinary neatness. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: Meets band +1; the line-width system and endpoint handling are unified, and optical alignment is meticulous. Quality improvement must be manifested as: the geometry, boundaries, and image proportions of different components coordinate with one another, remaining stable both up close and in overall viewing. Meets band +2; local craftsmanship is precise enough to serve as a presentation template, with no visible roughness. The finest connections, complex intersections, or irregular outlines are still handled adequately, and local craftsmanship does not damage the overall effect. Verify in this item's relevant complex parts: the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters. Task-specific object: Local craftsmanship such as arrow endpoints, line widths, frame cropping, and icon and text baselines should be clean; diagram connection relationships must not be distorted by misaligned endpoints or cropping. Difficult-case and inspection basis: Zoom in to check whether all arrow endpoints land on the target boxes/nodes, and whether line widths are systematic. Check whether image cropping cuts into chart labels or formulas. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. Fine construction resolves a real perceptual ambiguity at the task’s most demanding local junction or component, beyond spotless alignment and uniform polish. Identify the optical or geometric benefit visible at the intended scale; more detail or tiny changes with no perceptible benefit do not qualify. Apply the exceptional benefit specifically to: Local craftsmanship such as arrow endpoints, line widths, frame cropping, and icon and text baselines should be clean; diagram connection relationships must not be distorted by misaligned endpoints or cropping. In-task verification cases for that benefit: Zoom in to check whether all arrow endpoints land on the target boxes/nodes, and whether line widths are systematic. Check whether image cropping cuts into chart labels or formulas. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 6,
    "benchmark_source": "vas"
  },
  {
    "id": "V7",
    "dimension": "Visuals",
    "criterion": "Cross-page visible identity and content consistency.\nWhat is measured: Across the six pages, title styles, headers/footers, color usage, and figure-caption styles present the same report identity; when the same chart or symbol is reused across pages, its style and meaning are consistent.\nScope of observation: 6-page English PPTX and PNG, adapting the author's poster into an oral presentation for a research group.\nBasis for observation: Check side by side whether the six pages' title positions, font sizes, footers, page numbers, and color usage are consistent. Check whether figures, symbols, or annotation styles reused across pages remain consistent.\nScoring boundary: The stable correspondence of the same object, identity, markers, styles, and content across pages, states, and carriers; evaluate visual inheritance.\nHigh-band verification: Check only within this item's applicable scope the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters; complex states or the densest content remain consistent across all necessary carriers; identity, layout, and qualifications are all stably traceable.\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.\nCheck recurring objects in the actual slides and required exports; do not demand additional export formats.\nJudge negative visual grades by substantive harm and extent within this criterion, not a prerequisite of functional unusability. Harmless style differences are not defects; retain the original concrete defect examples.",
    "answers": {
      "-3": "Severe failure: This item's core understanding or use goal cannot be completed. Specific manifestation: the pages look like different templates stitched together, and identity is unrecognizable. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-2": "Clearly unacceptable: This item's main area or key process is persistently impaired. Specific manifestation: multiple inconsistencies in title/footer/color usage interrupt cross-page reading. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-1": "Below basic requirements: This item has a clear local substantive defect. Specific manifestation: individual pages drift in style and require manual unification. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "0": "Meets basic requirements: Cross-page styles are consistent, and it is evident they belong to the same report; this item's necessary requirements are fully met, with no substantive defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: Meets band 0; the title and footer system is stable, and reused figure styles are consistent. Quality improvement must be manifested as: repeated objects or corresponding carriers follow clear visual rules, reducing the burden of re-identification. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: Meets band +1; cross-page identity is unified and changes in a controlled way with content type, and the meaning of reused elements is clear. Quality improvement must be manifested as: state changes, repeated components, and necessary exports correspond to one another; content variation is distinguished while identity does not drift. Meets band +2; cross-page visual identity is mature and consistent, and stylistic variation also serves the argumentative hierarchy. Complex states or the densest content remain consistent across all necessary carriers; identity, layout, and qualifications are all stably traceable. Verify in this item's relevant complex parts: the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters. Task-specific object: Across the six pages, title styles, headers/footers, color usage, and figure-caption styles present the same report identity; when the same chart or symbol is reused across pages, its style and meaning are consistent. Difficult-case and inspection basis: Check side by side whether the six pages' title positions, font sizes, footers, page numbers, and color usage are consistent. Check whether figures, symbols, or annotation styles reused across pages remain consistent. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. A deliberately designed recurring visible identity enables recognition across the task’s genuinely different views or presentations with materially less re-identification effort than fully refined consistency alone. Name the recognition benefit and verify it across actual delivered contexts, without requiring nonexistent extra versions. Apply the exceptional benefit specifically to: Across the six pages, title styles, headers/footers, color usage, and figure-caption styles present the same report identity; when the same chart or symbol is reused across pages, its style and meaning are consistent. In-task verification cases for that benefit: Check side by side whether the six pages' title positions, font sizes, footers, page numbers, and color usage are consistent. Check whether figures, symbols, or annotation styles reused across pages remain consistent. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 3,
    "benchmark_source": "vas"
  },
  {
    "id": "V8",
    "dimension": "Visuals",
    "criterion": "Reading path and presentation rhythm of the static page sequence.\nWhat is measured: This item is a static PPTX; no new animation is required; the reading path and information density of the six pages should support the oral presentation rhythm: while staying on a single page, one can obtain claim, evidence, and qualification in sequence, and the page order does not cause repeated looking back; only static visual scanning/attention shifts/reading rhythm are evaluated, whether semantics connect between pages goes to O3, and real motion experience is not inferred from static screenshots.\nScope of observation: 6-page English PPTX and PNG, adapting the author's poster into an oral presentation for a research group.\nBasis for observation: Simulate page-by-page oral presentation, recording each page's first and second landing points and whether looking back is needed. Check whether the density of theory and experiments in the page order forms a tellable rhythm variation.\nScoring boundary: Eye movement between information units and content revelation, dwell, motion, transitions; evaluate scanning and temporal rhythm.\nHigh-band verification: Check only within this item's applicable scope the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters; the densest content or most complex handoff can still be followed comfortably, and the overall and detailed rhythm of the complete reading or playback cycle are consistent.\nApplicable boundary of this item (applies to all relevant entries, adds no new scoring item): Do not require adding animation, video, or dynamic transitions.\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.\nFor static slides, assess scanning and backtracking at presentation scale, not motion effects.\nJudge negative visual grades by substantive harm and extent within this criterion, not a prerequisite of functional unusability. Harmless style differences are not defects; retain the original concrete defect examples.",
    "answers": {
      "-3": "Severe failure: This item's core understanding or use goal cannot be completed. Specific manifestation: page information density and page order prevent the presentation from forming a rhythm, and the audience is frequently lost. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-2": "Clearly unacceptable: This item's main area or key process is persistently impaired. Specific manifestation: reading paths on multiple pages are chaotic or information is overloaded, forcing the oral presentation to skip around. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-1": "Below basic requirements: This item has a clear local substantive defect. Specific manifestation: the reading order on individual pages is not intuitive, and the presentation rhythm is interrupted. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "0": "Meets basic requirements: The page order allows smooth presentation, and the single-page reading path holds; this item's necessary requirements are fully met, with no substantive defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: Meets band 0; page information density has variation, and it is clear what to present first. Quality improvement must be manifested as: one class of scanning path, dwell, or handoff arrangements makes necessary information easier to finish reading and follow. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: Meets band +1; the page order and within-page reading paths coordinate with the oral rhythm, and theoretical and experimental sections have well-paced tension and relaxation. Quality improvement must be manifested as: eye-movement or temporal arrangements coordinate with content difficulty, with no unnecessary jumping back or waiting between entering, understanding, and leaving. Meets band +2; the overall rhythm is mature, and even static pages can form a clear presentation curve without additional animation to compensate. The densest content or most complex handoff can still be followed comfortably, and the overall and detailed rhythm of the complete reading or playback cycle are consistent. Verify in this item's relevant complex parts: the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters. Task-specific object: This item is a static PPTX; no new animation is required; the reading path and information density of the six pages should support the oral presentation rhythm: while staying on a single page, one can obtain claim, evidence, and qualification in sequence, and the page order does not cause repeated looking back; only static visual scanning/attention shifts/reading rhythm are evaluated, whether semantics connect between pages goes to O3, and real motion experience is not inferred from static screenshots. Difficult-case and inspection basis: Simulate page-by-page oral presentation, recording each page's first and second landing points and whether looking back is needed. Check whether the density of theory and experiments in the page order forms a tellable rhythm variation. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. A non-obvious reading or temporal organization lets the audience orient, inspect and continue with unusually little backtracking or missed understanding. Identify the timing or scanning tension solved beyond consistently smooth ordinary presentation; add no motion or interaction to a static task. Apply the exceptional benefit specifically to: This item is a static PPTX; no new animation is required; the reading path and information density of the six pages should support the oral presentation rhythm: while staying on a single page, one can obtain claim, evidence, and qualification in sequence, and the page order does not cause repeated looking back; only static visual scanning/attention shifts/reading rhythm are evaluated, whether semantics connect between pages goes to O3, and real motion experience is not inferred from static screenshots. In-task verification cases for that benefit: Simulate page-by-page oral presentation, recording each page's first and second landing points and whether looking back is needed. Check whether the density of theory and experiments in the page order forms a tellable rhythm variation. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 3,
    "benchmark_source": "vas"
  },
  {
    "id": "V9",
    "dimension": "Visuals",
    "criterion": "Independent aesthetics of the color scheme and image background colors.\nWhat is measured: The color scheme itself should suit academic projection: the combination of hue/lightness/saturation is restrained, accent colors are used for evidence rather than decoration, and the image background colors of poster excerpts coordinate with the page background.\nScope of observation: 6-page English PPTX and PNG, adapting the author's poster into an oral presentation for a research group.\nBasis for observation: Check the relationship among the page's main color, accent colors, and image background colors at projection brightness. Check whether accent colors fall only on key evidence or conclusions.\nScoring boundary: Hue, lightness, saturation × area, accent colors, and the aesthetic relationship between images/background colors; demonstrate quality goals separately from V3/V4/V5/V7.\nObservation elements: Hue compatibility and warm/cool relationships. Lightness hierarchy and color center of gravity. Saturation × area proportions. The number, weight, and restraint of accent colors. Integration of images, graphics, panels, and background colors.\nHigh-band verification: Check only within this item's applicable scope the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters; in the hardest-to-coordinate images, dense regions, or varying states, the relationships among lightness, hue, and area remain unified, clear, and restrained.\nScoring boundary: V9 independently evaluates the aesthetic harmony and refinement of the color scheme; reading impairment caused by text contrast goes to V3, mismatches between color and object meaning go to V4, and cross-page color-code identity drift goes to V7. Cross-item attribution is made only when different functional impairments are demonstrated. V5 cannot merely repeat V9's color-scheme quality assessment, and V9 cannot directly award the professional band merely for \"fitting the scenario.\"\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.\nJudge negative visual grades by substantive harm and extent within this criterion, not a prerequisite of functional unusability. Harmless style differences are not defects; retain the original concrete defect examples.",
    "answers": {
      "-3": "Severe failure: This item's core understanding or use goal cannot be completed. Specific manifestation: the color scheme is glaring or the contrast is chaotic, making it hard to look at for long under projection, and image background colors severely conflict with the page. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-2": "Clearly unacceptable: This item's main area or key process is persistently impaired. Specific manifestation: multiple color conflicts or misuse of accent colors, and the image background colors of poster excerpts are severed from the page. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-1": "Below basic requirements: This item has a clear local substantive defect. Specific manifestation: the color scheme is somewhat chaotic or image/text background colors are uncoordinated but do not block reading. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "0": "Meets basic requirements: The color scheme is normal and restrained, with no substantive discomfort, and image background colors can coexist with the page; this item's necessary requirements are fully met, with no substantive defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: Meets band 0; the color scheme has a unified logic, accent color use is justified, and image background colors are basically incorporated into the page. Quality improvement must be manifested as: pointing out how specific hue, lightness, or saturation-and-area relationships improve harmony, rather than merely listing color names. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: Meets band +1; the relationship between hue area and lightness has been designed, the image background colors of poster excerpts coordinate with the page background, and accent colors fall on evidence or conclusions. Quality improvement must be manifested as: the relationships among subject, background, images, and accent colors are coordinated, and color center of gravity and area control serve actual content. Meets band +2; the color scheme is mature and has an academic character, emphasis and background colors are precisely integrated, and it does not rely on flashiness or a fixed single color family. In the hardest-to-coordinate images, dense regions, or varying states, the relationships among lightness, hue, and area remain unified, clear, and restrained. Verify in this item's relevant complex parts: the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters. Task-specific object: The color scheme itself should suit academic projection: the combination of hue/lightness/saturation is restrained, accent colors are used for evidence rather than decoration, and the image background colors of poster excerpts coordinate with the page background. Difficult-case and inspection basis: Check the relationship among the page's main color, accent colors, and image background colors at projection brightness. Check whether accent colors fall only on key evidence or conclusions. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. Subtle hue, lightness, saturation and area relationships reconcile the hardest color-balance conflict with a distinctive, highly resolved visual character. Explain the perceptual gain beyond a flawlessly harmonious palette; semantic color correctness, fashion and personal liking do not establish exceptional color design. Apply the exceptional benefit specifically to: The color scheme itself should suit academic projection: the combination of hue/lightness/saturation is restrained, accent colors are used for evidence rather than decoration, and the image background colors of poster excerpts coordinate with the page background. In-task verification cases for that benefit: Check the relationship among the page's main color, accent colors, and image background colors at projection brightness. Check whether accent colors fall only on key evidence or conclusions. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 6,
    "benchmark_source": "vas"
  },
  {
    "id": "D1",
    "dimension": "Delivery",
    "criterion": "Openability and locatability of the required deliverables.\nWhat is measured: artifact/report.pptx must be a normally openable six-page English PPTX; the six slide PNGs, sources.md, and delivery.md must be named as specified by the prompt and located in artifact/, directly locatable and usable.\nScope of observation: 6-page English PPTX and PNG, adapting the author's poster into an oral presentation for a research group.\nBasis for observation: Open artifact/report.pptx and check the page count and English content. Check whether the six PNGs correspond one-to-one with the PPTX pages, and whether sources.md and delivery.md are in artifact/.\nScoring boundary: The openability, locatability, and direct usability of the required files; V7 evaluates the consistency of actually visible content, and file existence does not substitute for quality.\nHigh-band verification: Check the complete viewing flow from locating the main product to the necessary auxiliary materials according to this item's actual delivery scope; if one item is missing, judge the band only by actual impact, and do not require out-of-scope formats or extra files.\nScenario reference · medium (adds no new threshold): English PPTX (six pages from cover to conclusion), one PNG per page, sources.md, delivery.md, editable text and graphics.\nApplicable boundary of this item (applies to all relevant entries, adds no new scoring item): input/ and prompt.md are read-only originals and must not be modified; final deliverables are written to artifact/.\nApplicable boundary of this item (applies to all relevant entries, adds no new scoring item): Preserve the theorem's restrictions on scale/models, and the conclusion's dependence on the conjecture (including the 1-cycle vs 2-cycle conjecture).\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.",
    "answers": {
      "-3": "Severe failure: A core required deliverable is missing or, because of its own defects, cannot be opened/used as a whole, making the task's core use objective impossible to achieve; first rule out evaluation-tool and environment failures.",
      "-2": "Clearly unacceptable: Required content is missing on a large scale, or key opening/locating processes are persistently obstructed, requiring substantial remediation of the main use workflow; judge according to actual impact, not by mechanically assigning a tier based solely on a single discrepancy in file count.",
      "-1": "Below basic requirements: There is a clear, localized issue involving a missing deliverable, version identification, file correspondence, or opening instructions that affects necessary use under this criterion, while the remaining core content is usable.",
      "0": "Meets basic requirements: The six-page PPTX opens, and the PNGs and both md files are complete and in the correct locations; this item's necessary requirements are fully met, with no substantive defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: within the task-specified files and entry points, explicit role, sequence, or location cues eliminate one actual search or trial-and-error step; complete files and filenames that comply with the task statement alone qualify only for tier 0. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: the roles, sequence, and necessary usage instructions of required files are coordinated, and the viewing path from the main artifact to corresponding previews/supporting materials is streamlined and unambiguous; assess only the files actually required by the task statement. even the required files, versions, or states most likely to be confused can be located accurately, and the complete viewing workflow is smooth; actual opening and required correspondence verification support the judgment, without requiring additional navigation pages or explanatory documents. Task-specific object: artifact/report.pptx must be a normally openable six-page English PPTX; the six slide PNGs, sources.md, and delivery.md must be named as specified by the prompt and located in artifact/, directly locatable and usable. Difficult-case and inspection basis: Open artifact/report.pptx and check the page count and English content. Check whether the six PNGs correspond one-to-one with the PPTX pages, and whether sources.md and delivery.md are in artifact/. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. Within the required deliverables and entry points, an economical handoff removes a real selection, opening or first-use ambiguity that persists in a conventionally flawless delivery. Identify the materially reduced effort; extra files, packaging or software are not required. If no additional benefit is observable, +2 is the ceiling. Apply the exceptional benefit specifically to: artifact/report.pptx must be a normally openable six-page English PPTX; the six slide PNGs, sources.md, and delivery.md must be named as specified by the prompt and located in artifact/, directly locatable and usable. In-task verification cases for that benefit: Open artifact/report.pptx and check the page count and English content. Check whether the six PNGs correspond one-to-one with the PPTX pages, and whether sources.md and delivery.md are in artifact/. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 4,
    "benchmark_source": "vas"
  },
  {
    "id": "D2",
    "dimension": "Delivery",
    "criterion": "Editable structure and non-rasterization.\nWhat is measured: Key explanations must remain editable text and graphics; poster excerpts may be used, but entire pages must not be rasterized; text, formulas, and self-drawn diagrams in the PPTX should be editable, and offline resources must not depend on the network.\nScope of observation: 6-page English PPTX and PNG, adapting the author's poster into an oral presentation for a research group.\nBasis for observation: In the PPTX, click through each page's key explanatory text and diagrams to confirm they are not full-page images. Check whether there are network-dependent fonts/image links, and confirm it opens offline.\nScoring boundary: The editable structure, resources, and material organization of the objects required by the prompt; no engineering scripts or extra tests are required.\nHigh-band verification: Verify key explanatory text, formulas, and self-drawn diagrams, allowing poster evidence excerpts in the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters' structure, resources, and editing entry points; do not modify submitted files, and do not require unrequested generation scripts or extra documents.\nApplicable boundary of this item (applies to all relevant entries, adds no new scoring item): Do not require providing TEX source or other format branches not listed in the prompt.\nApplicable boundary of this item (applies to all relevant entries, adds no new scoring item): Poster excerpts may be used, but key explanations must remain editable text and graphics and must not be rasterized page by page.\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.",
    "answers": {
      "-3": "Severe failure: This item's core understanding or use goal cannot be completed. Specific manifestation: entire pages consist of poster screenshots, and key explanations are not editable. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-2": "Clearly unacceptable: This item's main area or key process is persistently impaired. Specific manifestation: key text or diagrams on multiple pages are rasterized and cannot be edited. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-1": "Below basic requirements: This item has a clear local substantive defect. Specific manifestation: individual key explanations exist in image form but can be locally remedied. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "0": "Meets basic requirements: Key explanatory text, formulas, and self-drawn diagrams, with poster evidence excerpts allowed, are complete and editable as specified by the prompt, and necessary resources can be resolved; the boundary between native content and allowed raster assets is correct, and there is no dependence on the author's private missing paths; this item's necessary requirements are fully met, with no substantive defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: Meets band 0; in key explanatory text, formulas, and self-drawn diagrams, with poster evidence excerpts allowed, object naming, grouping, or data relationships have specific arrangements that make locating/local maintenance of one class of necessary content more direct than ordinary editability; mere existence of objects is not sufficient for a positive score. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: Meets band +1; in key explanatory text, formulas, and self-drawn diagrams, with poster evidence excerpts allowed, semantic grouping, resources, and editing entry points are coordinated, complex objects can be located, and local maintenance does not disturb unrelated content; no extra general-purpose engine or refactoring is required. Meets band +2; in the pages with the densest method/theorem conditions and the experimental conclusion recovery; not requiring listing all theorem parameters, among the corresponding necessary objects, both complex structures and simple parts are clearly maintainable, and resource relationships and original correct content are stable; based on native structure and prompt-specified build checks, without modifying the submission or adding out-of-scope experiments. Task-specific object: Key explanations must remain editable text and graphics; poster excerpts may be used, but entire pages must not be rasterized; text, formulas, and self-drawn diagrams in the PPTX should be editable, and offline resources must not depend on the network. Difficult-case and inspection basis: In the PPTX, click through each page's key explanatory text and diagrams to confirm they are not full-page images. Check whether there are network-dependent fonts/image links, and confirm it opens offline. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. The native editable organization reveals a meaningful relationship among task objects so directly that a targeted maintenance step requires materially less structural reconstruction than in a flawlessly editable conventional artifact. Show this through native evidence, without requiring an extra refactoring, unrequested source code or speculative editing claims. Apply the exceptional benefit specifically to: Key explanations must remain editable text and graphics; poster excerpts may be used, but entire pages must not be rasterized; text, formulas, and self-drawn diagrams in the PPTX should be editable, and offline resources must not depend on the network. In-task verification cases for that benefit: In the PPTX, click through each page's key explanatory text and diagrams to confirm they are not full-page images. Check whether there are network-dependent fonts/image links, and confirm it opens offline. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 3,
    "benchmark_source": "vas"
  },
  {
    "id": "D3",
    "dimension": "Delivery",
    "criterion": "Verifiability of sources, versions, doubts, and delivery notes.\nWhat is measured: sources.md must record sources, versions, and figure provenance, and record poster/paper version differences, ambiguities, or verification results (including the doubt about the Result threshold symbol in the upper right); delivery.md must explain the actual delivered content; the notes must be consistent with the artifacts and verifiable.\nScope of observation: 6-page English PPTX and PNG, adapting the author's poster into an oral presentation for a research group.\nBasis for observation: Check whether sources.md lists the poster/paper versions, figure provenance, and at least one version difference or doubt. Check whether the files described in delivery.md are consistent with the actual contents of artifact/.\nScoring boundary: The verifiability of required sources, calculations/changes/captures or usage notes; select according to this item's actual requirements, and do not require all document types.\nHigh-band verification: Check only within this item's applicable scope the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters; even the most easily confused sources, conditions, or changes can be checked along the notes, and the complete verification path is concise with no gaps or exaggeration.\nApplicable boundary of this item (applies to all relevant entries, adds no new scoring item): Preserve the poster/paper version differences and the doubt about checking the Result threshold symbol in the upper right, and record them in sources.md; do not make rewriting the suspected symbol mandatory.\nRevised grade boundary: the original high-grade descriptions are +2 quality obligations, not sufficient evidence for +3. +2 covers the entire applicable scope; +3 additionally requires the exceptional benefit specified in answers[\"3\"]. Levels inherit quality obligations, not the shortcomings explaining lower grades.",
    "answers": {
      "-3": "Severe failure: This item's core understanding or use goal cannot be completed. Specific manifestation: source and version notes are missing, and figure provenance cannot be verified. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-2": "Clearly unacceptable: This item's main area or key process is persistently impaired. Specific manifestation: notes clearly do not match the artifacts, or key version differences/doubts are not recorded. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "-1": "Below basic requirements: This item has a clear local substantive defect. Specific manifestation: notes exist but lack one of version differences or figure provenance. Judged by the above actual impact; harmless minor differences do not trigger this band.",
      "0": "Meets basic requirements: sources.md and delivery.md are complete, and sources and versions are discernible; this item's necessary requirements are fully met, with no substantive defects, and performance is ordinary.",
      "1": "Accomplished but not yet fully refined: meets tier 0 and shows a substantive observed benefit beyond ordinary correctness, completeness or readability through the following: Meets band 0; sources, versions, figure provenance, and discovered doubts are all recorded and correspond to actual files. Quality improvement must be manifested as: at least one class of key basis has a precise location and corresponds to an actual object, reducing the need to search through vague sources. In reason explain the concrete benefit and the observable limitation keeping it below +2. Do not invent a defect to force this grade; missing required observation must be acquired first.",
      "2": "Fully refined within scope: meets tier 0 and achieves complete quality in every applicable part, not only an attractive local example. Retain and jointly verify these quality obligations: Meets band +1; version differences, conjecture dependencies, and symbol doubts are clearly recorded, and the delivery notes facilitate reuse and verification. Quality improvement must be manifested as: claims, materials, and the prompt's required notes correspond to one another, so that one can both trace the basis and know where that basis is used. Meets band +2; source and delivery notes are rigorous and transparent, and the handling of uncertainty can serve as a verification example. Even the most easily confused sources, conditions, or changes can be checked along the notes, and the complete verification path is concise with no gaps or exaggeration. Verify in this item's relevant complex parts: the pages with the densest method/theorem conditions and the experimental conclusion recovery; do not require listing all theorem parameters. Task-specific object: sources.md must record sources, versions, and figure provenance, and record poster/paper version differences, ambiguities, or verification results (including the doubt about the Result threshold symbol in the upper right); delivery.md must explain the actual delivered content; the notes must be consistent with the artifacts and verifiable. Difficult-case and inspection basis: Check whether sources.md lists the poster/paper versions, figure provenance, and at least one version difference or doubt. Check whether the files described in delivery.md are consistent with the actual contents of artifact/. Within this criterion and the stated use conditions, there is no identifiable substantive weakness or feasible task-relevant improvement of material benefit. Establish full applicable coverage and actively inspect the strongest plausible shortcoming; finding no error, a single successful example or an author assertion does not prove full refinement.",
      "3": "Exceptional beyond conventional full refinement: first fully meet +2. An economical provenance/handoff organization makes a difficult fact, calculation or asset relationship directly auditable, reducing reconstruction beyond a complete accurate conventional source list. Identify the actual trace and saved ambiguity, without requiring new evidence or new documentation beyond the task. Apply the exceptional benefit specifically to: sources.md must record sources, versions, and figure provenance, and record poster/paper version differences, ambiguities, or verification results (including the doubt about the Result threshold symbol in the upper right); delivery.md must explain the actual delivered content; the notes must be consistent with the artifacts and verifiable. In-task verification cases for that benefit: Check whether sources.md lists the poster/paper versions, figure provenance, and at least one version difference or doubt. Check whether the files described in delivery.md are consistent with the actual contents of artifact/. Correct completion of these cases is only a foundation, not evidence of exceptional benefit. In reason identify the non-obvious choice, real difficulty and substantial benefit beyond an equally correct, complete and polished conventional solution, with actual evidence. No comparison artifact or unique design is required. Complexity, extra features, ornament, praise or repetition of +2 strengths is insufficient. Without demonstrated extra benefit the ceiling is +2; no score quota applies."
    },
    "weight": 3,
    "benchmark_source": "vas"
  }
]