Show pure Kinex RLE evaluation progress; retain native VLA and historical Kinex evidence

hard/app.js

@@ -1,6 +1,12 @@
 function rleProtocolNote() {
  const n=data.instruction_protocol_notice;if(!n)return '';
- return `<aside class="native-note" data-rle-protocol><span class="note-symbol">i</span><div><strong>Pure Kinex · RLE revised instructions</strong><p>GPT-6 Astra / high with rosbridge; ${number(n.selected_kinex)} selected episodes. No VLA extension or device. Existing VLA results retain original native instructions, admission rules and execution. Different instruction and control protocols prevent a paired causal comparison.</p><div class="inline-links">${external(evidenceURL('research/kinex-rle-20261104/REPORT.md'),'RLE evaluation report')}${external(evidenceURL('research/kinex-rle-20261104/plan.json'),'Frozen task selection & protocol')}${n.historical_kinex?external(resultURL('attempt-history.json'),'Previous native Kinex results'):''}</div></div></aside>`;
+ const seeds=n.seed_overrides?`<p data-plus-seed-summary><strong>Plus-only seed overrides for adjust_bottle:</strong> ${Object.entries(n.seed_overrides).map(([config,seed])=>esc(config.replace('demo_',''))+' = '+seed).join('; ')}. Each is the minimum initializable seed from ascending zero-model checks. Easy and Hard retain their established seeds, including validated seed 0.</p>`:'';
+ return `<aside class="native-note" data-rle-protocol><span class="note-symbol">i</span><div><strong>Pure Kinex · RLE revised instructions</strong><p>GPT-6 Astra / high with rosbridge; ${number(n.selected_kinex)} selected episodes. No VLA extension or device. Existing VLA results retain original native instructions, admission rules and execution. Different instruction and control protocols prevent a paired causal comparison.</p>${seeds}<div class="inline-links">${external(evidenceURL('research/kinex-rle-20261104/REPORT.md'),'RLE evaluation report')}${external(evidenceURL('research/kinex-rle-20261104/plan.json'),'Frozen task selection & protocol')}${n.historical_kinex?external(resultURL('attempt-history.json'),'Previous native Kinex results'):''}</div></div></aside>`;
+}
+function plusSeedNote(task,variant,ep) {
+ const selection=variant==='kinex'?(ep?.measured_protocol?.seed_selection||task.planned_protocol_by_variant?.kinex?.seed_selection):null;
+ if(!selection)return '';
+ return `<aside class="native-note" data-plus-seed-note><span class="note-symbol">i</span><div><strong>Plus-only initialization seed: ${number(selection.selected_seed)}</strong><p>Authored seed 0 could not initialize this Plus task/configuration. Examined seeds: ${selection.examined_seeds.map(number).join(', ')}. The first initializable seed was selected before model use, without expert-success filtering. One formal episode is scheduled. Easy and Hard keep their established seeds, including previously validated seed 0.</p><div class="inline-links">${external(selection.evidence,'Seed selection and preflight evidence')}</div></div></aside>`;
 }
 'use strict';
 const plannedProtocol = (task, variant) => task.planned_protocol_by_variant?.[variant] || task.planned_protocol;
@@ -238,7 +244,7 @@ function taskPage(key,variant,section='overview') {
   sessionState={filter:'all',query:'',count:40,view:'timeline'};
   const taskList=plannedTasks(variant),index=taskList.indexOf(task),history=data.attempt_history.filter(h=>h.task_key===key&&h.variant===variant);
   document.title=`${task.title} · ${variantFor(variant).agent} — Programmable VLA`;
-  $('#main').innerHTML=`<div class="breadcrumbs"><a data-task-return href="${esc(taskListReturn?.hash || '#tasks?task='+encodeURIComponent(key))}">← Back to tasks</a><span>/</span><a href="#runs/${variant}?family=${task.family}">${familyName(task.family)}</a><span>/</span><code>${key}</code></div><div class="task-heading"><div><div class="eyebrow">${familyName(task.family)} · ${key}</div><h1>${esc(task.title)}</h1><p><code>${esc(task.native_id)}</code></p></div><div class="task-arrows">${index?`<a class="button" href="${taskRoute(taskList[index-1].key,variant)}" aria-label="Previous task">←</a>`:''}${index<taskList.length-1?`<a class="button" href="${taskRoute(taskList[index+1].key,variant)}" aria-label="Next task">→</a>`:''}</div></div>${budgetNote(task)}${variantTabs(variant,key,section)}<nav class="detail-nav" aria-label="Episode sections">${[['overview','Overview',null],['session','Session',ep?.session_counts?.visible_events],['tools','Tool calls',ep?.call_activity?.model_tool_calls],['policy','VLA input/output',ep?.vla_calls],['artifacts','Artifacts',ep?.resources?.length],['evidence','Evidence',null],['attempts','Previous attempts',history.length]].filter(([k])=>(k!=='attempts'||history.length)&&(k!=='policy'||variantFor(variant).agent!=='Kinex')).map(([k,l,n])=>!ep&&!['overview','attempts'].includes(k)?`<span class="disabled-tab" aria-disabled="true">${l}</span>`:`<a href="${taskRoute(key,variant,k)}" class="${k===section?'active':''}" ${k===section?'aria-current="page"':''}>${l}${n!=null?`<span>${number(n)}</span>`:''}</a>`).join('')}<span class="detail-model">${variantFor(variant).model} / ${variantFor(variant).effort}</span></nav><div id="detail-content">${section==='overview'?overviewDetail(task,ep):section==='session'?sessionShell(ep):section==='tools'?toolsShell(ep):section==='policy'?policyShell(ep):section==='artifacts'?artifactsPanel(ep):section==='evidence'?evidencePanel(task,ep):historyPanel(history,task)}</div>`;
+  $('#main').innerHTML=`<div class="breadcrumbs"><a data-task-return href="${esc(taskListReturn?.hash || '#tasks?task='+encodeURIComponent(key))}">← Back to tasks</a><span>/</span><a href="#runs/${variant}?family=${task.family}">${familyName(task.family)}</a><span>/</span><code>${key}</code></div><div class="task-heading"><div><div class="eyebrow">${familyName(task.family)} · ${key}</div><h1>${esc(task.title)}</h1><p><code>${esc(task.native_id)}</code></p></div><div class="task-arrows">${index?`<a class="button" href="${taskRoute(taskList[index-1].key,variant)}" aria-label="Previous task">←</a>`:''}${index<taskList.length-1?`<a class="button" href="${taskRoute(taskList[index+1].key,variant)}" aria-label="Next task">→</a>`:''}</div></div>${budgetNote(task)}${plusSeedNote(task,variant,ep)}${variantTabs(variant,key,section)}<nav class="detail-nav" aria-label="Episode sections">${[['overview','Overview',null],['session','Session',ep?.session_counts?.visible_events],['tools','Tool calls',ep?.call_activity?.model_tool_calls],['policy','VLA input/output',ep?.vla_calls],['artifacts','Artifacts',ep?.resources?.length],['evidence','Evidence',null],['attempts','Previous attempts',history.length]].filter(([k])=>(k!=='attempts'||history.length)&&(k!=='policy'||variantFor(variant).agent!=='Kinex')).map(([k,l,n])=>!ep&&!['overview','attempts'].includes(k)?`<span class="disabled-tab" aria-disabled="true">${l}</span>`:`<a href="${taskRoute(key,variant,k)}" class="${k===section?'active':''}" ${k===section?'aria-current="page"':''}>${l}${n!=null?`<span>${number(n)}</span>`:''}</a>`).join('')}<span class="detail-model">${variantFor(variant).model} / ${variantFor(variant).effort}</span></nav><div id="detail-content">${section==='overview'?overviewDetail(task,ep):section==='session'?sessionShell(ep):section==='tools'?toolsShell(ep):section==='policy'?policyShell(ep):section==='artifacts'?artifactsPanel(ep):section==='evidence'?evidencePanel(task,ep):historyPanel(history,task)}</div>`;
   setupVideo();
   if(section==='session')loadSession(ep,epoch);
   if(section==='tools')loadTools(ep,epoch);

hard/data-source.json

@@ -1,26 +1,26 @@
 {
   "schema": "programvla/dataset-source/1",
   "dataset": "RLE-Bench/programmable-vla",
-  "revision": "09872377e2d2825af2c6925b835dba4c68ace410",
-  "evidence_root": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/",
+  "revision": "a99c740fc04f65cedda40c128bbae3e373d41c5e",
+  "evidence_root": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/",
   "files": {
     "data.json": {
-      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/research/kinex-rle-20261104/panels/hard/data.json"
+      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/research/kinex-rle-20261104/panels/hard/data.json"
     },
     "cost-data.json": {
-      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/research/kinex-rle-20261104/panels/hard/cost-data.json"
+      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/research/kinex-rle-20261104/panels/hard/cost-data.json"
     },
     "measured-instructions.json": {
-      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/research/kinex-rle-20261104/panels/hard/measured-instructions.json"
+      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/research/kinex-rle-20261104/panels/hard/measured-instructions.json"
     },
     "sources.json": {
-      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/research/kinex-rle-20261104/panels/hard/sources.json"
+      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/research/kinex-rle-20261104/panels/hard/sources.json"
     },
     "instruction-audit.json": {
-      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/research/kinex-rle-20261104/panels/hard/instruction-audit.json"
+      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/research/kinex-rle-20261104/panels/hard/instruction-audit.json"
     },
     "attempt-history.json": {
-      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/research/kinex-rle-20261104/panels/hard/attempt-history.json"
+      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/research/kinex-rle-20261104/panels/hard/attempt-history.json"
     }
   }
 }

hard/index.html

@@ -18,9 +18,9 @@
 <a class="skip" href="#main">Skip to content</a>
 <header class="site-header"><div class="header-inner"><a class="brand" href="#overview" aria-label="Programmable VLA home"><img src="assets/kinex-logo.svg" width="34" height="34" alt="Kinex"><span>programmable<span class="brand-divider">/</span><span class="brand-benchmark">vla</span></span></a><nav aria-label="Main navigation"><a href="../plus/index.html">Plus</a><a href="../index.html#overview">Easy</a><a href="#overview" data-nav="overview">Hard</a><a href="#tasks" data-nav="tasks">Explore tasks</a><a href="#instructions" data-nav="instructions">Instructions audit</a><a href="#methodology" data-nav="methodology">Protocol</a></nav><a class="header-link" href="https://huggingface.co/spaces/RLE-Bench/programmable-vla" target="_blank" rel="noopener">Hugging Face <span aria-hidden="true">↗</span></a></div></header>
 <main id="main" tabindex="-1"><div class="loading"><span class="spinner"></span>Loading benchmark results…</div></main>
-<footer class="site-footer"><div><img src="assets/kinex-logo.svg" width="22" height="22" alt=""><strong>Programmable VLA</strong><span class="footer-org">by RLE-Bench</span></div><div><a data-result-file="data.json" href="https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/research/kinex-rle-20261104/panels/hard/data.json" target="_blank">Hard results JSON</a><a data-result-file="sources.json" href="https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/b07d88579523eeeefdf7ad2ca14ee65340af5d4e/hard/sources.json" target="_blank">Data sources</a><a href="https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/b07d88579523eeeefdf7ad2ca14ee65340af5d4e/manifest.json" target="_blank">Manifest</a><span id="updated"></span></div></footer>
+<footer class="site-footer"><div><img src="assets/kinex-logo.svg" width="22" height="22" alt=""><strong>Programmable VLA</strong><span class="footer-org">by RLE-Bench</span></div><div><a data-result-file="data.json" href="https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/research/kinex-rle-20261104/panels/hard/data.json" target="_blank">Hard results JSON</a><a data-result-file="sources.json" href="https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/b07d88579523eeeefdf7ad2ca14ee65340af5d4e/hard/sources.json" target="_blank">Data sources</a><a href="https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/b07d88579523eeeefdf7ad2ca14ee65340af5d4e/manifest.json" target="_blank">Manifest</a><span id="updated"></span></div></footer>
 <dialog id="image-viewer" aria-labelledby="image-caption"><div class="viewer-toolbar"><span id="image-caption"></span><a id="image-original" target="_blank" rel="noopener">Open original ↗</a><button class="icon-button" id="close-viewer" aria-label="Close image">×</button></div><div class="viewer-image"><button id="previous-image" class="viewer-arrow" aria-label="Previous image">←</button><img id="large-image" alt="Recorded agent observation"><button id="next-image" class="viewer-arrow" aria-label="Next image">→</button></div><p>Recorded agent observation · Use ← → to browse · Esc to close</p></dialog>
 <div id="toast" role="status" aria-live="polite"></div>
-<noscript><p>This site needs JavaScript. <a data-result-file="data.json" href="https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/research/kinex-rle-20261104/panels/hard/data.json">Download the complete results index.</a></p></noscript>
+<noscript><p>This site needs JavaScript. <a data-result-file="data.json" href="https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/research/kinex-rle-20261104/panels/hard/data.json">Download the complete results index.</a></p></noscript>
 </body>
 </html>

manifest.json

@@ -99,8 +99,8 @@
       "sha256": "a53108594a23b740fbb08623f6fffdda1d73608b552988a9ffda9f0944ae66a1"
     },
     "hard/app.js": {
-      "bytes": 96176,
-      "sha256": "65b425cdd73418e4e7e307237e24509d55170a06f633ca5fc64b860ee0bd9019"
+      "bytes": 97458,
+      "sha256": "4f9e18564115bc5d6e660491fa58e40e72d5ed41aed7c1cb23d49fe295025d90"
     },
     "hard/icon.svg": {
       "bytes": 725,
@@ -112,7 +112,7 @@
     },
     "hard/index.html": {
       "bytes": 3630,
-      "sha256": "3a90c03c1ce36ef0d3e5983927ae25e68f15a2c6bcdc84dcb0771f3e629a5109"
+      "sha256": "13a49ae2c2b496144000024195ef386a19016de45dfbec6e31537b001ee70344"
     },
     "hard/policy.css": {
       "bytes": 1067,
@@ -124,11 +124,11 @@
     },
     "hard/data-source.json": {
       "bytes": 1587,
-      "sha256": "7020856190548dc51bbc5e0347d6f6c8f37c5a21ca21d462012d5596b914a297"
+      "sha256": "a3f5876d89e6a253934b22fb922973e8eeeec74b203c3d9ebaab5bb18fecfbc9"
     },
     "plus/app.js": {
-      "bytes": 98374,
-      "sha256": "c208f79c4c755ec65fa14d36081de3482e23b69861df53b5aaa1c84890b228f3"
+      "bytes": 99656,
+      "sha256": "9755e4bed00adc61e918ca955976454cf90142420c36f5bec7cfe929e6b1ea2f"
     },
     "plus/icon.svg": {
       "bytes": 725,
@@ -140,7 +140,7 @@
     },
     "plus/index.html": {
       "bytes": 3639,
-      "sha256": "2d8bc51fc4c6fda8ec06ac9cbc4e4a30b8d81a478b006eb29f648e04e9c5d246"
+      "sha256": "64d33d40f57c75d74e34bad657f057e2f229409afe7ef0d42cd4e75a42edacae"
     },
     "plus/policy.css": {
       "bytes": 1067,
@@ -152,7 +152,7 @@
     },
     "plus/data-source.json": {
       "bytes": 1587,
-      "sha256": "ab66f393d90c8fb365271b3ab91093422b8b299c9ee6e13391ba80f879a5f3c0"
+      "sha256": "93b73c07119774436c222b8623b17bd2aadc2c87565b4597535728dd642d780f"
     },
     "instruction-audit/sources/task03-manifest.yaml": {
       "bytes": 5747,
@@ -255,7 +255,7 @@
   },
   "kinex_rle_dataset": {
     "repo": "RLE-Bench/programmable-vla",
-    "revision": "09872377e2d2825af2c6925b835dba4c68ace410",
+    "revision": "a99c740fc04f65cedda40c128bbae3e373d41c5e",
     "prefix": "research/kinex-rle-20261104",
     "scope": "Pure Kinex, RLE revised instructions; Hard 50 selected, Plus 24 selected",
     "complete": false

plus/app.js

@@ -1,6 +1,12 @@
 function rleProtocolNote() {
  const n=data.instruction_protocol_notice;if(!n)return '';
- return `<aside class="native-note" data-rle-protocol><span class="note-symbol">i</span><div><strong>Pure Kinex · RLE revised instructions</strong><p>GPT-6 Astra / high with rosbridge; ${number(n.selected_kinex)} selected episodes. No VLA extension or device. Existing VLA results retain original native instructions, admission rules and execution. Different instruction and control protocols prevent a paired causal comparison.</p><div class="inline-links">${external(evidenceURL('research/kinex-rle-20261104/REPORT.md'),'RLE evaluation report')}${external(evidenceURL('research/kinex-rle-20261104/plan.json'),'Frozen task selection & protocol')}${n.historical_kinex?external(resultURL('attempt-history.json'),'Previous native Kinex results'):''}</div></div></aside>`;
+ const seeds=n.seed_overrides?`<p data-plus-seed-summary><strong>Plus-only seed overrides for adjust_bottle:</strong> ${Object.entries(n.seed_overrides).map(([config,seed])=>esc(config.replace('demo_',''))+' = '+seed).join('; ')}. Each is the minimum initializable seed from ascending zero-model checks. Easy and Hard retain their established seeds, including validated seed 0.</p>`:'';
+ return `<aside class="native-note" data-rle-protocol><span class="note-symbol">i</span><div><strong>Pure Kinex · RLE revised instructions</strong><p>GPT-6 Astra / high with rosbridge; ${number(n.selected_kinex)} selected episodes. No VLA extension or device. Existing VLA results retain original native instructions, admission rules and execution. Different instruction and control protocols prevent a paired causal comparison.</p>${seeds}<div class="inline-links">${external(evidenceURL('research/kinex-rle-20261104/REPORT.md'),'RLE evaluation report')}${external(evidenceURL('research/kinex-rle-20261104/plan.json'),'Frozen task selection & protocol')}${n.historical_kinex?external(resultURL('attempt-history.json'),'Previous native Kinex results'):''}</div></div></aside>`;
+}
+function plusSeedNote(task,variant,ep) {
+ const selection=variant==='kinex'?(ep?.measured_protocol?.seed_selection||task.planned_protocol_by_variant?.kinex?.seed_selection):null;
+ if(!selection)return '';
+ return `<aside class="native-note" data-plus-seed-note><span class="note-symbol">i</span><div><strong>Plus-only initialization seed: ${number(selection.selected_seed)}</strong><p>Authored seed 0 could not initialize this Plus task/configuration. Examined seeds: ${selection.examined_seeds.map(number).join(', ')}. The first initializable seed was selected before model use, without expert-success filtering. One formal episode is scheduled. Easy and Hard keep their established seeds, including previously validated seed 0.</p><div class="inline-links">${external(selection.evidence,'Seed selection and preflight evidence')}</div></div></aside>`;
 }
 'use strict';
 const plannedProtocol = (task, variant) => task.planned_protocol_by_variant?.[variant] || task.planned_protocol;
@@ -238,7 +244,7 @@ function taskPage(key,variant,section='overview') {
   sessionState={filter:'all',query:'',count:40,view:'timeline'};
   const taskList=plannedTasks(variant),index=taskList.indexOf(task),history=data.attempt_history.filter(h=>h.task_key===key&&h.variant===variant);
   document.title=`${task.title} · ${variantFor(variant).agent} — Programmable VLA`;
-  $('#main').innerHTML=`<div class="breadcrumbs"><a data-task-return href="${esc(taskListReturn?.hash || '#tasks?task='+encodeURIComponent(key))}">← Back to tasks</a><span>/</span><a href="#runs/${variant}?family=${task.family}">${familyName(task.family)}</a><span>/</span><code>${key}</code></div><div class="task-heading"><div><div class="eyebrow">${familyName(task.family)} · ${key}</div><h1>${esc(task.title)}</h1><p><code>${esc(task.native_id)}</code></p></div><div class="task-arrows">${index?`<a class="button" href="${taskRoute(taskList[index-1].key,variant)}" aria-label="Previous task">←</a>`:''}${index<taskList.length-1?`<a class="button" href="${taskRoute(taskList[index+1].key,variant)}" aria-label="Next task">→</a>`:''}</div></div>${budgetNote(task)}${variantTabs(variant,key,section)}<nav class="detail-nav" aria-label="Episode sections">${[['overview','Overview',null],['session','Session',ep?.session_counts?.visible_events],['tools','Tool calls',ep?.call_activity?.model_tool_calls],['policy','VLA input/output',ep?.vla_calls],['artifacts','Artifacts',ep?.resources?.length],['evidence','Evidence',null],['attempts','Previous attempts',history.length]].filter(([k])=>(k!=='attempts'||history.length)&&(k!=='policy'||variantFor(variant).agent!=='Kinex')).map(([k,l,n])=>!ep&&!['overview','attempts'].includes(k)?`<span class="disabled-tab" aria-disabled="true">${l}</span>`:`<a href="${taskRoute(key,variant,k)}" class="${k===section?'active':''}" ${k===section?'aria-current="page"':''}>${l}${n!=null?`<span>${number(n)}</span>`:''}</a>`).join('')}<span class="detail-model">${variantFor(variant).model} / ${variantFor(variant).effort}</span></nav><div id="detail-content">${section==='overview'?overviewDetail(task,ep):section==='session'?sessionShell(ep):section==='tools'?toolsShell(ep):section==='policy'?policyShell(ep):section==='artifacts'?artifactsPanel(ep):section==='evidence'?evidencePanel(task,ep):historyPanel(history,task)}</div>`;
+  $('#main').innerHTML=`<div class="breadcrumbs"><a data-task-return href="${esc(taskListReturn?.hash || '#tasks?task='+encodeURIComponent(key))}">← Back to tasks</a><span>/</span><a href="#runs/${variant}?family=${task.family}">${familyName(task.family)}</a><span>/</span><code>${key}</code></div><div class="task-heading"><div><div class="eyebrow">${familyName(task.family)} · ${key}</div><h1>${esc(task.title)}</h1><p><code>${esc(task.native_id)}</code></p></div><div class="task-arrows">${index?`<a class="button" href="${taskRoute(taskList[index-1].key,variant)}" aria-label="Previous task">←</a>`:''}${index<taskList.length-1?`<a class="button" href="${taskRoute(taskList[index+1].key,variant)}" aria-label="Next task">→</a>`:''}</div></div>${budgetNote(task)}${plusSeedNote(task,variant,ep)}${variantTabs(variant,key,section)}<nav class="detail-nav" aria-label="Episode sections">${[['overview','Overview',null],['session','Session',ep?.session_counts?.visible_events],['tools','Tool calls',ep?.call_activity?.model_tool_calls],['policy','VLA input/output',ep?.vla_calls],['artifacts','Artifacts',ep?.resources?.length],['evidence','Evidence',null],['attempts','Previous attempts',history.length]].filter(([k])=>(k!=='attempts'||history.length)&&(k!=='policy'||variantFor(variant).agent!=='Kinex')).map(([k,l,n])=>!ep&&!['overview','attempts'].includes(k)?`<span class="disabled-tab" aria-disabled="true">${l}</span>`:`<a href="${taskRoute(key,variant,k)}" class="${k===section?'active':''}" ${k===section?'aria-current="page"':''}>${l}${n!=null?`<span>${number(n)}</span>`:''}</a>`).join('')}<span class="detail-model">${variantFor(variant).model} / ${variantFor(variant).effort}</span></nav><div id="detail-content">${section==='overview'?overviewDetail(task,ep):section==='session'?sessionShell(ep):section==='tools'?toolsShell(ep):section==='policy'?policyShell(ep):section==='artifacts'?artifactsPanel(ep):section==='evidence'?evidencePanel(task,ep):historyPanel(history,task)}</div>`;
   setupVideo();
   if(section==='session')loadSession(ep,epoch);
   if(section==='tools')loadTools(ep,epoch);

plus/data-source.json

@@ -1,26 +1,26 @@
 {
   "schema": "programvla/dataset-source/1",
   "dataset": "RLE-Bench/programmable-vla",
-  "revision": "09872377e2d2825af2c6925b835dba4c68ace410",
-  "evidence_root": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/",
+  "revision": "a99c740fc04f65cedda40c128bbae3e373d41c5e",
+  "evidence_root": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/",
   "files": {
     "data.json": {
-      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/research/kinex-rle-20261104/panels/plus/data.json"
+      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/research/kinex-rle-20261104/panels/plus/data.json"
     },
     "cost-data.json": {
-      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/research/kinex-rle-20261104/panels/plus/cost-data.json"
+      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/research/kinex-rle-20261104/panels/plus/cost-data.json"
     },
     "measured-instructions.json": {
-      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/research/kinex-rle-20261104/panels/plus/measured-instructions.json"
+      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/research/kinex-rle-20261104/panels/plus/measured-instructions.json"
     },
     "sources.json": {
-      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/research/kinex-rle-20261104/panels/plus/sources.json"
+      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/research/kinex-rle-20261104/panels/plus/sources.json"
     },
     "instruction-audit.json": {
-      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/research/kinex-rle-20261104/panels/plus/instruction-audit.json"
+      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/research/kinex-rle-20261104/panels/plus/instruction-audit.json"
     },
     "attempt-history.json": {
-      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/research/kinex-rle-20261104/panels/plus/attempt-history.json"
+      "url": "https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/research/kinex-rle-20261104/panels/plus/attempt-history.json"
     }
   }
 }

plus/index.html

@@ -18,9 +18,9 @@
 <a class="skip" href="#main">Skip to content</a>
 <header class="site-header"><div class="header-inner"><a class="brand" href="#overview" aria-label="Programmable VLA home"><img src="assets/kinex-logo.svg" width="34" height="34" alt="Kinex"><span>programmable<span class="brand-divider">/</span><span class="brand-benchmark">vla</span></span></a><nav aria-label="Main navigation"><a href="../index.html#overview">Easy</a><a href="../hard/index.html#overview">Hard</a><a href="#overview" data-nav="overview">Plus</a><a href="#tasks" data-nav="tasks">Explore tasks</a><a href="#instructions" data-nav="instructions">Instructions audit</a><a href="#methodology" data-nav="methodology">Protocol</a></nav><a class="header-link" href="https://huggingface.co/spaces/RLE-Bench/programmable-vla" target="_blank" rel="noopener">Hugging Face <span aria-hidden="true">↗</span></a></div></header>
 <main id="main" tabindex="-1"><div class="loading"><span class="spinner"></span>Loading benchmark results…</div></main>
-<footer class="site-footer"><div><img src="assets/kinex-logo.svg" width="22" height="22" alt=""><strong>Programmable VLA</strong><span class="footer-org">by RLE-Bench</span></div><div><a data-result-file="data.json" href="https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/research/kinex-rle-20261104/panels/plus/data.json" target="_blank">Plus results JSON</a><a data-result-file="sources.json" href="https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/b07d88579523eeeefdf7ad2ca14ee65340af5d4e/hard/sources.json" target="_blank">Data sources</a><a href="https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/b2d67a67dd2f9d8e12c1f8db13ba8de309739883/manifest.json" target="_blank">Manifest</a><span id="updated"></span></div></footer>
+<footer class="site-footer"><div><img src="assets/kinex-logo.svg" width="22" height="22" alt=""><strong>Programmable VLA</strong><span class="footer-org">by RLE-Bench</span></div><div><a data-result-file="data.json" href="https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/research/kinex-rle-20261104/panels/plus/data.json" target="_blank">Plus results JSON</a><a data-result-file="sources.json" href="https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/b07d88579523eeeefdf7ad2ca14ee65340af5d4e/hard/sources.json" target="_blank">Data sources</a><a href="https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/b2d67a67dd2f9d8e12c1f8db13ba8de309739883/manifest.json" target="_blank">Manifest</a><span id="updated"></span></div></footer>
 <dialog id="image-viewer" aria-labelledby="image-caption"><div class="viewer-toolbar"><span id="image-caption"></span><a id="image-original" target="_blank" rel="noopener">Open original ↗</a><button class="icon-button" id="close-viewer" aria-label="Close image">×</button></div><div class="viewer-image"><button id="previous-image" class="viewer-arrow" aria-label="Previous image">←</button><img id="large-image" alt="Recorded agent observation"><button id="next-image" class="viewer-arrow" aria-label="Next image">→</button></div><p>Recorded agent observation · Use ← → to browse · Esc to close</p></dialog>
 <div id="toast" role="status" aria-live="polite"></div>
-<noscript><p>This site needs JavaScript. <a data-result-file="data.json" href="https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/09872377e2d2825af2c6925b835dba4c68ace410/research/kinex-rle-20261104/panels/plus/data.json">Download the complete results index.</a></p></noscript>
+<noscript><p>This site needs JavaScript. <a data-result-file="data.json" href="https://huggingface.co/datasets/RLE-Bench/programmable-vla/resolve/a99c740fc04f65cedda40c128bbae3e373d41c5e/research/kinex-rle-20261104/panels/plus/data.json">Download the complete results index.</a></p></noscript>
 </body>
 </html>