d3-edge-CoreAI / prompt_reference.json
devilyouwei's picture
Add locally converted text-only D3 Core AI model and full-duplex validation
0690c04 verified
Raw History Blame Contribute Delete
59.5 kB
{
"short_cn": {
"records": [
{
"id": "duplex_stop_01",
"expected": "interrupt",
"choice": "interrupt",
"probabilities": [
0.9308630228042603,
0.03596342355012894,
0.03317355364561081
],
"seconds": 0.2473604999249801,
"tokens": 218,
"reference_probabilities": [
0.9297863841056824,
0.03624925762414932,
0.033964335918426514
],
"reference_choice": "interrupt",
"max_probability_error": 0.0010766386985778809,
"reference_seconds": 1.3486362090334296,
"choice_matches": true
},
{
"id": "duplex_stop_02",
"expected": "interrupt",
"choice": "interrupt",
"probabilities": [
0.6084311604499817,
0.1538606435060501,
0.2377082109451294
],
"seconds": 0.07908791594672948,
"tokens": 218,
"reference_probabilities": [
0.5986561179161072,
0.15632180869579315,
0.24502210319042206
],
"reference_choice": "interrupt",
"max_probability_error": 0.009775042533874512,
"reference_seconds": 1.5567700000246987,
"choice_matches": true
},
{
"id": "duplex_stop_03",
"expected": "interrupt",
"choice": "interrupt",
"probabilities": [
0.9178251624107361,
0.02585407719016075,
0.05632082000374794
],
"seconds": 0.07947220804635435,
"tokens": 219,
"reference_probabilities": [
0.9217900037765503,
0.024559389799833298,
0.0536506213247776
],
"reference_choice": "interrupt",
"max_probability_error": 0.003964841365814209,
"reference_seconds": 1.5973578330595046,
"choice_matches": true
},
{
"id": "duplex_stop_04",
"expected": "interrupt",
"choice": "interrupt",
"probabilities": [
0.9476383924484253,
0.029724864289164543,
0.022636760026216507
],
"seconds": 0.07833625003695488,
"tokens": 220,
"reference_probabilities": [
0.9495176076889038,
0.028405454009771347,
0.022076912224292755
],
"reference_choice": "interrupt",
"max_probability_error": 0.0018792152404785156,
"reference_seconds": 2.027006958029233,
"choice_matches": true
},
{
"id": "duplex_correct_01",
"expected": "interrupt",
"choice": "interrupt",
"probabilities": [
0.7275057435035706,
0.07282276451587677,
0.19967152178287506
],
"seconds": 0.0755601660348475,
"tokens": 222,
"reference_probabilities": [
0.7318935990333557,
0.07304492592811584,
0.1950615495443344
],
"reference_choice": "interrupt",
"max_probability_error": 0.004609972238540649,
"reference_seconds": 1.9489428750239313,
"choice_matches": true
},
{
"id": "duplex_correct_02",
"expected": "interrupt",
"choice": "interrupt",
"probabilities": [
0.8345935344696045,
0.057409752160310745,
0.10799670219421387
],
"seconds": 0.0829312918940559,
"tokens": 222,
"reference_probabilities": [
0.8331737518310547,
0.05937865749001503,
0.10744760930538177
],
"reference_choice": "interrupt",
"max_probability_error": 0.0019689053297042847,
"reference_seconds": 1.9444509589811787,
"choice_matches": true
},
{
"id": "duplex_correct_03",
"expected": "interrupt",
"choice": "interrupt",
"probabilities": [
0.6642118692398071,
0.06811445951461792,
0.26767367124557495
],
"seconds": 0.07992016698699445,
"tokens": 221,
"reference_probabilities": [
0.6618426442146301,
0.06853331625461578,
0.2696240544319153
],
"reference_choice": "interrupt",
"max_probability_error": 0.002369225025177002,
"reference_seconds": 1.9678149169776589,
"choice_matches": true
},
{
"id": "duplex_switch_01",
"expected": "interrupt",
"choice": "interrupt",
"probabilities": [
0.7909741997718811,
0.09334924072027206,
0.11567656695842743
],
"seconds": 0.08040650002658367,
"tokens": 223,
"reference_probabilities": [
0.7888679504394531,
0.0943671464920044,
0.11676487326622009
],
"reference_choice": "interrupt",
"max_probability_error": 0.0021062493324279785,
"reference_seconds": 1.9147198750870302,
"choice_matches": true
},
{
"id": "duplex_urgent_01",
"expected": "interrupt",
"choice": "interrupt",
"probabilities": [
0.8331965804100037,
0.11239644140005112,
0.05440700426697731
],
"seconds": 0.0784068750217557,
"tokens": 223,
"reference_probabilities": [
0.8334749937057495,
0.11307955533266068,
0.053445473313331604
],
"reference_choice": "interrupt",
"max_probability_error": 0.0009615309536457062,
"reference_seconds": 1.930482874973677,
"choice_matches": true
},
{
"id": "duplex_ack_01",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.05335284024477005,
0.7110395431518555,
0.2356075942516327
],
"seconds": 0.07608508295379579,
"tokens": 216,
"reference_probabilities": [
0.055138103663921356,
0.7180846929550171,
0.22677722573280334
],
"reference_choice": "discard",
"max_probability_error": 0.008830368518829346,
"reference_seconds": 1.827593500027433,
"choice_matches": true
},
{
"id": "duplex_ack_02",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.0421481616795063,
0.6526758670806885,
0.3051760196685791
],
"seconds": 0.07729579193983227,
"tokens": 218,
"reference_probabilities": [
0.04397556185722351,
0.6295878887176514,
0.32643648982048035
],
"reference_choice": "discard",
"max_probability_error": 0.02308797836303711,
"reference_seconds": 1.8134827500907704,
"choice_matches": true
},
{
"id": "duplex_ack_03",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.040575310587882996,
0.6564949750900269,
0.30292972922325134
],
"seconds": 0.07996416697278619,
"tokens": 217,
"reference_probabilities": [
0.04088408499956131,
0.6473596692085266,
0.3117562234401703
],
"reference_choice": "discard",
"max_probability_error": 0.009135305881500244,
"reference_seconds": 1.8543335000285879,
"choice_matches": true
},
{
"id": "duplex_ack_04",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.011969413608312607,
0.9218174815177917,
0.06621308624744415
],
"seconds": 0.07615158404223621,
"tokens": 216,
"reference_probabilities": [
0.012495238333940506,
0.9165955185890198,
0.07090925425291061
],
"reference_choice": "discard",
"max_probability_error": 0.005221962928771973,
"reference_seconds": 2.1056112080113962,
"choice_matches": true
},
{
"id": "duplex_continue_01",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.04026160016655922,
0.8019729852676392,
0.15776535868644714
],
"seconds": 0.07585625001229346,
"tokens": 218,
"reference_probabilities": [
0.04062916710972786,
0.7997573614120483,
0.15961341559886932
],
"reference_choice": "discard",
"max_probability_error": 0.0022156238555908203,
"reference_seconds": 1.7713218339486048,
"choice_matches": true
},
{
"id": "duplex_negation_01",
"expected": "discard",
"choice": "interrupt",
"probabilities": [
0.6325174570083618,
0.2781596779823303,
0.08932279795408249
],
"seconds": 0.07802729099057615,
"tokens": 220,
"reference_probabilities": [
0.6228411197662354,
0.28723961114883423,
0.08991922438144684
],
"reference_choice": "interrupt",
"max_probability_error": 0.009676337242126465,
"reference_seconds": 1.948609041981399,
"choice_matches": true
},
{
"id": "duplex_negation_02",
"expected": "discard",
"choice": "interrupt",
"probabilities": [
0.8180356621742249,
0.07483810931444168,
0.10712619870901108
],
"seconds": 0.08067808300256729,
"tokens": 218,
"reference_probabilities": [
0.8170282244682312,
0.0774642676115036,
0.10550743341445923
],
"reference_choice": "interrupt",
"max_probability_error": 0.00262615829706192,
"reference_seconds": 1.8800741248996928,
"choice_matches": true
},
{
"id": "duplex_noise_01",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.16634003818035126,
0.7751500010490417,
0.05850996449589729
],
"seconds": 0.07809220894705504,
"tokens": 218,
"reference_probabilities": [
0.1708865612745285,
0.7685359120368958,
0.060577549040317535
],
"reference_choice": "discard",
"max_probability_error": 0.006614089012145996,
"reference_seconds": 1.8496827089693397,
"choice_matches": true
},
{
"id": "duplex_noise_02",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.140708327293396,
0.5931727290153503,
0.2661188840866089
],
"seconds": 0.07563637499697506,
"tokens": 214,
"reference_probabilities": [
0.14067772030830383,
0.6005666255950928,
0.258755624294281
],
"reference_choice": "discard",
"max_probability_error": 0.007393896579742432,
"reference_seconds": 2.0071601249510422,
"choice_matches": true
},
{
"id": "duplex_echo_01",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.06657762825489044,
0.8454098701477051,
0.08801253885030746
],
"seconds": 0.07592029101215303,
"tokens": 220,
"reference_probabilities": [
0.06640980392694473,
0.8431832790374756,
0.0904068872332573
],
"reference_choice": "discard",
"max_probability_error": 0.002394348382949829,
"reference_seconds": 1.9474582499824464,
"choice_matches": true
},
{
"id": "duplex_other_01",
"expected": "discard",
"choice": "interrupt",
"probabilities": [
0.6339613795280457,
0.18493230640888214,
0.1811063289642334
],
"seconds": 0.07726554095279425,
"tokens": 221,
"reference_probabilities": [
0.6387590169906616,
0.1780863255262375,
0.1831546574831009
],
"reference_choice": "interrupt",
"max_probability_error": 0.006845980882644653,
"reference_seconds": 1.921368166920729,
"choice_matches": true
},
{
"id": "duplex_queue_01",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.4120011627674103,
0.08334042876958847,
0.5046584010124207
],
"seconds": 0.07846620806958526,
"tokens": 221,
"reference_probabilities": [
0.41622427105903625,
0.08575785905122757,
0.49801793694496155
],
"reference_choice": "enqueue",
"max_probability_error": 0.0066404640674591064,
"reference_seconds": 2.770340499933809,
"choice_matches": true
},
{
"id": "duplex_queue_02",
"expected": "enqueue",
"choice": "interrupt",
"probabilities": [
0.4853450059890747,
0.04944157600402832,
0.46521347761154175
],
"seconds": 0.07490887504536659,
"tokens": 222,
"reference_probabilities": [
0.48474135994911194,
0.05074368417263031,
0.4645150303840637
],
"reference_choice": "interrupt",
"max_probability_error": 0.0013021081686019897,
"reference_seconds": 2.5253281670156866,
"choice_matches": true
},
{
"id": "duplex_queue_03",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.20899155735969543,
0.124051034450531,
0.6669573783874512
],
"seconds": 0.07843541703186929,
"tokens": 221,
"reference_probabilities": [
0.2090408354997635,
0.12236136943101883,
0.6685976982116699
],
"reference_choice": "enqueue",
"max_probability_error": 0.0016896650195121765,
"reference_seconds": 1.929755000048317,
"choice_matches": true
},
{
"id": "duplex_queue_04",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.09898572415113449,
0.14187583327293396,
0.7591385245323181
],
"seconds": 0.0787995409918949,
"tokens": 220,
"reference_probabilities": [
0.10049959272146225,
0.13858738541603088,
0.7609130144119263
],
"reference_choice": "enqueue",
"max_probability_error": 0.003288447856903076,
"reference_seconds": 1.9884955000597984,
"choice_matches": true
},
{
"id": "duplex_queue_05",
"expected": "enqueue",
"choice": "interrupt",
"probabilities": [
0.5718603730201721,
0.149464413523674,
0.2786753177642822
],
"seconds": 0.07880450005177408,
"tokens": 222,
"reference_probabilities": [
0.5775824189186096,
0.15099485218524933,
0.27142271399497986
],
"reference_choice": "interrupt",
"max_probability_error": 0.007252603769302368,
"reference_seconds": 2.1081489589996636,
"choice_matches": true
},
{
"id": "duplex_queue_06",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.08183945715427399,
0.1113441213965416,
0.8068164587020874
],
"seconds": 0.07932108291424811,
"tokens": 222,
"reference_probabilities": [
0.08436998724937439,
0.11129157245159149,
0.8043384552001953
],
"reference_choice": "enqueue",
"max_probability_error": 0.002530530095100403,
"reference_seconds": 2.4768615000648424,
"choice_matches": true
},
{
"id": "duplex_queue_07",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.20892539620399475,
0.1285468339920044,
0.6625277400016785
],
"seconds": 0.0786395410541445,
"tokens": 218,
"reference_probabilities": [
0.20809990167617798,
0.12942783534526825,
0.6624722480773926
],
"reference_choice": "enqueue",
"max_probability_error": 0.000881001353263855,
"reference_seconds": 1.6376230830792338,
"choice_matches": true
},
{
"id": "duplex_idle_01",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.3379806578159332,
0.19012808799743652,
0.4718911945819855
],
"seconds": 0.08041737508028746,
"tokens": 220,
"reference_probabilities": [
0.34042972326278687,
0.18757516145706177,
0.47199517488479614
],
"reference_choice": "enqueue",
"max_probability_error": 0.002552926540374756,
"reference_seconds": 4.18124191602692,
"choice_matches": true
},
{
"id": "duplex_idle_02",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.22359061241149902,
0.26534196734428406,
0.5110673904418945
],
"seconds": 0.08483033289667219,
"tokens": 214,
"reference_probabilities": [
0.2355312705039978,
0.27137330174446106,
0.49309539794921875
],
"reference_choice": "enqueue",
"max_probability_error": 0.01797199249267578,
"reference_seconds": 2.817324834060855,
"choice_matches": true
},
{
"id": "duplex_idle_03",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.2232709527015686,
0.14533931016921997,
0.6313897371292114
],
"seconds": 0.08540104201529175,
"tokens": 216,
"reference_probabilities": [
0.22581645846366882,
0.14200592041015625,
0.6321775913238525
],
"reference_choice": "enqueue",
"max_probability_error": 0.0033333897590637207,
"reference_seconds": 1.4794647499220446,
"choice_matches": true
},
{
"id": "duplex_idle_04",
"expected": "enqueue",
"choice": "discard",
"probabilities": [
0.025776008144021034,
0.7816474437713623,
0.19257661700248718
],
"seconds": 0.08535249996930361,
"tokens": 217,
"reference_probabilities": [
0.026971980929374695,
0.7835420966148376,
0.18948595225811005
],
"reference_choice": "discard",
"max_probability_error": 0.0030906647443771362,
"reference_seconds": 1.4521212499821559,
"choice_matches": true
},
{
"id": "duplex_idle_05",
"expected": "enqueue",
"choice": "interrupt",
"probabilities": [
0.4170467257499695,
0.3020874261856079,
0.2808658480644226
],
"seconds": 0.0853517500218004,
"tokens": 217,
"reference_probabilities": [
0.40342554450035095,
0.3200090229511261,
0.27656546235084534
],
"reference_choice": "interrupt",
"max_probability_error": 0.01792159676551819,
"reference_seconds": 1.4486252079950646,
"choice_matches": true
},
{
"id": "duplex_idle_06",
"expected": "enqueue",
"choice": "discard",
"probabilities": [
0.32025524973869324,
0.40157341957092285,
0.2781713306903839
],
"seconds": 0.08389929099939764,
"tokens": 218,
"reference_probabilities": [
0.3254876732826233,
0.39759308099746704,
0.27691924571990967
],
"reference_choice": "discard",
"max_probability_error": 0.005232423543930054,
"reference_seconds": 1.4733430419582874,
"choice_matches": true
},
{
"id": "duplex_idle_07",
"expected": "discard",
"choice": "interrupt",
"probabilities": [
0.595440149307251,
0.18141186237335205,
0.22314797341823578
],
"seconds": 0.07983308401890099,
"tokens": 221,
"reference_probabilities": [
0.5790136456489563,
0.19464221596717834,
0.22634416818618774
],
"reference_choice": "interrupt",
"max_probability_error": 0.016426503658294678,
"reference_seconds": 1.4898792499443516,
"choice_matches": true
},
{
"id": "duplex_idle_08",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.023530151695013046,
0.7448784708976746,
0.2315913885831833
],
"seconds": 0.07574620901141316,
"tokens": 217,
"reference_probabilities": [
0.022543322294950485,
0.7479900121688843,
0.22946661710739136
],
"reference_choice": "discard",
"max_probability_error": 0.003111541271209717,
"reference_seconds": 1.5462311250157654,
"choice_matches": true
},
{
"id": "duplex_answer_01",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.36920368671417236,
0.21816276013851166,
0.4126335680484772
],
"seconds": 0.07867716706823558,
"tokens": 218,
"reference_probabilities": [
0.3760320544242859,
0.2168779969215393,
0.40709003806114197
],
"reference_choice": "enqueue",
"max_probability_error": 0.006828367710113525,
"reference_seconds": 1.410918749985285,
"choice_matches": true
},
{
"id": "duplex_answer_02",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.3135741651058197,
0.19335414469242096,
0.49307164549827576
],
"seconds": 0.07869200001005083,
"tokens": 222,
"reference_probabilities": [
0.316304087638855,
0.19665490090847015,
0.48704105615615845
],
"reference_choice": "enqueue",
"max_probability_error": 0.00603058934211731,
"reference_seconds": 1.8347832500003278,
"choice_matches": true
},
{
"id": "duplex_answer_03",
"expected": "enqueue",
"choice": "discard",
"probabilities": [
0.35142919421195984,
0.3655969202518463,
0.28297391533851624
],
"seconds": 0.0753501660656184,
"tokens": 219,
"reference_probabilities": [
0.3477523922920227,
0.3722348213195801,
0.2800127863883972
],
"reference_choice": "discard",
"max_probability_error": 0.006637901067733765,
"reference_seconds": 1.5008892079349607,
"choice_matches": true
},
{
"id": "duplex_question_stop",
"expected": "enqueue",
"choice": "interrupt",
"probabilities": [
0.8747069239616394,
0.0360749326646328,
0.0892181396484375
],
"seconds": 0.07888579193968326,
"tokens": 222,
"reference_probabilities": [
0.8738645911216736,
0.0369415320456028,
0.0891939178109169
],
"reference_choice": "interrupt",
"max_probability_error": 0.0008665993809700012,
"reference_seconds": 1.4881447920342907,
"choice_matches": true
},
{
"id": "duplex_quoted_stop",
"expected": "enqueue",
"choice": "interrupt",
"probabilities": [
0.5751999020576477,
0.07048548012971878,
0.3543146252632141
],
"seconds": 0.07968554203398526,
"tokens": 224,
"reference_probabilities": [
0.5706831216812134,
0.0702967494726181,
0.3590201437473297
],
"reference_choice": "interrupt",
"max_probability_error": 0.004705518484115601,
"reference_seconds": 1.5156835409579799,
"choice_matches": true
},
{
"id": "duplex_mid_ack_request",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.28059422969818115,
0.11433308571577072,
0.6050726771354675
],
"seconds": 0.08052141696680337,
"tokens": 222,
"reference_probabilities": [
0.28669923543930054,
0.1158427894115448,
0.597458004951477
],
"reference_choice": "enqueue",
"max_probability_error": 0.0076146721839904785,
"reference_seconds": 1.585977791924961,
"choice_matches": true
},
{
"id": "duplex_immediate_01",
"expected": "interrupt",
"choice": "interrupt",
"probabilities": [
0.6138381958007812,
0.24859386682510376,
0.1375679224729538
],
"seconds": 0.07979370898101479,
"tokens": 221,
"reference_probabilities": [
0.6091759204864502,
0.25300347805023193,
0.13782061636447906
],
"reference_choice": "interrupt",
"max_probability_error": 0.004662275314331055,
"reference_seconds": 1.654788792016916,
"choice_matches": true
},
{
"id": "duplex_wait_01",
"expected": "interrupt",
"choice": "interrupt",
"probabilities": [
0.6828376650810242,
0.1435249298810959,
0.17363739013671875
],
"seconds": 0.07526308309752494,
"tokens": 221,
"reference_probabilities": [
0.6814437508583069,
0.14467670023441315,
0.17387954890727997
],
"reference_choice": "interrupt",
"max_probability_error": 0.0013939142227172852,
"reference_seconds": 1.440822875010781,
"choice_matches": true
},
{
"id": "duplex_wait_02",
"expected": "enqueue",
"choice": "interrupt",
"probabilities": [
0.8187282681465149,
0.07863655686378479,
0.10263523459434509
],
"seconds": 0.08320104097947478,
"tokens": 222,
"reference_probabilities": [
0.824877142906189,
0.07670652866363525,
0.09841634333133698
],
"reference_choice": "interrupt",
"max_probability_error": 0.006148874759674072,
"reference_seconds": 1.4709639159264043,
"choice_matches": true
},
{
"id": "duplex_blank_01",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.23855049908161163,
0.5664229989051819,
0.19502650201320648
],
"seconds": 0.08054837491363287,
"tokens": 213,
"reference_probabilities": [
0.23826876282691956,
0.5627028346061707,
0.19902841746807098
],
"reference_choice": "discard",
"max_probability_error": 0.004001915454864502,
"reference_seconds": 1.4517487920820713,
"choice_matches": true
},
{
"id": "duplex_idle_correction",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.3666907846927643,
0.09975643455982208,
0.5335528254508972
],
"seconds": 0.07898704102262855,
"tokens": 222,
"reference_probabilities": [
0.3673265874385834,
0.10558851063251495,
0.5270849466323853
],
"reference_choice": "enqueue",
"max_probability_error": 0.006467878818511963,
"reference_seconds": 1.567459666985087,
"choice_matches": true
},
{
"id": "duplex_queue_constraint",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.23049047589302063,
0.0817626342177391,
0.6877469420433044
],
"seconds": 0.08358375006355345,
"tokens": 222,
"reference_probabilities": [
0.23331351578235626,
0.08226655423641205,
0.6844199299812317
],
"reference_choice": "enqueue",
"max_probability_error": 0.003327012062072754,
"reference_seconds": 1.5755315830465406,
"choice_matches": true
},
{
"id": "duplex_late_request",
"expected": "enqueue",
"choice": "interrupt",
"probabilities": [
0.6512891054153442,
0.1271398365497589,
0.22157108783721924
],
"seconds": 0.08454970794264227,
"tokens": 224,
"reference_probabilities": [
0.6481541395187378,
0.12712408602237701,
0.22472184896469116
],
"reference_choice": "interrupt",
"max_probability_error": 0.003150761127471924,
"reference_seconds": 1.6315815409179777,
"choice_matches": true
}
],
"choices_matched": 48,
"cases": 48,
"max_probability_error": 0.02308797836303711,
"reference_business_correct": 34
},
"short_en": {
"records": [
{
"id": "duplex_stop_01",
"expected": "interrupt",
"choice": "interrupt",
"probabilities": [
0.6268212795257568,
0.19353297352790833,
0.17964576184749603
],
"seconds": 0.08400687505491078,
"tokens": 228,
"reference_probabilities": [
0.6213389039039612,
0.19729731976985931,
0.1813637763261795
],
"reference_choice": "interrupt",
"max_probability_error": 0.005482375621795654,
"reference_seconds": 1.552941375062801,
"choice_matches": true
},
{
"id": "duplex_stop_02",
"expected": "interrupt",
"choice": "enqueue",
"probabilities": [
0.10337332636117935,
0.363031804561615,
0.5335948467254639
],
"seconds": 0.08210300002247095,
"tokens": 228,
"reference_probabilities": [
0.10244514048099518,
0.37220701575279236,
0.5253477692604065
],
"reference_choice": "enqueue",
"max_probability_error": 0.009175211191177368,
"reference_seconds": 1.5801174161024392,
"choice_matches": true
},
{
"id": "duplex_stop_03",
"expected": "interrupt",
"choice": "interrupt",
"probabilities": [
0.5057383179664612,
0.3377652168273926,
0.15649652481079102
],
"seconds": 0.08048545802012086,
"tokens": 229,
"reference_probabilities": [
0.5172834396362305,
0.33390575647354126,
0.14881081879138947
],
"reference_choice": "interrupt",
"max_probability_error": 0.011545121669769287,
"reference_seconds": 1.5210892499890178,
"choice_matches": true
},
{
"id": "duplex_stop_04",
"expected": "interrupt",
"choice": "interrupt",
"probabilities": [
0.6581788659095764,
0.23939505219459534,
0.10242605209350586
],
"seconds": 0.08602541696745902,
"tokens": 230,
"reference_probabilities": [
0.6520472168922424,
0.24444346129894257,
0.1035093441605568
],
"reference_choice": "interrupt",
"max_probability_error": 0.006131649017333984,
"reference_seconds": 1.5384657080285251,
"choice_matches": true
},
{
"id": "duplex_correct_01",
"expected": "interrupt",
"choice": "enqueue",
"probabilities": [
0.18660441040992737,
0.3201097846031189,
0.4932858347892761
],
"seconds": 0.08271174994297326,
"tokens": 232,
"reference_probabilities": [
0.18579913675785065,
0.3210013508796692,
0.4931994676589966
],
"reference_choice": "enqueue",
"max_probability_error": 0.000891566276550293,
"reference_seconds": 1.6162409170065075,
"choice_matches": true
},
{
"id": "duplex_correct_02",
"expected": "interrupt",
"choice": "interrupt",
"probabilities": [
0.35816383361816406,
0.3007599413394928,
0.3410762548446655
],
"seconds": 0.08182208298239857,
"tokens": 232,
"reference_probabilities": [
0.36502936482429504,
0.29951760172843933,
0.33545297384262085
],
"reference_choice": "interrupt",
"max_probability_error": 0.0068655312061309814,
"reference_seconds": 1.6218001670204103,
"choice_matches": true
},
{
"id": "duplex_correct_03",
"expected": "interrupt",
"choice": "enqueue",
"probabilities": [
0.16310663521289825,
0.22092485427856445,
0.6159685254096985
],
"seconds": 0.08308241597842425,
"tokens": 231,
"reference_probabilities": [
0.16249701380729675,
0.21792727708816528,
0.6195757389068604
],
"reference_choice": "enqueue",
"max_probability_error": 0.0036072134971618652,
"reference_seconds": 1.636168249999173,
"choice_matches": true
},
{
"id": "duplex_switch_01",
"expected": "interrupt",
"choice": "enqueue",
"probabilities": [
0.12074483931064606,
0.3092057704925537,
0.5700494050979614
],
"seconds": 0.08102087501902133,
"tokens": 233,
"reference_probabilities": [
0.11796737462282181,
0.312320351600647,
0.569712221622467
],
"reference_choice": "enqueue",
"max_probability_error": 0.0031145811080932617,
"reference_seconds": 1.5371713329805061,
"choice_matches": true
},
{
"id": "duplex_urgent_01",
"expected": "interrupt",
"choice": "interrupt",
"probabilities": [
0.7929849624633789,
0.12692339718341827,
0.08009161800146103
],
"seconds": 0.08281858405098319,
"tokens": 233,
"reference_probabilities": [
0.788321852684021,
0.12926553189754486,
0.08241260796785355
],
"reference_choice": "interrupt",
"max_probability_error": 0.00466310977935791,
"reference_seconds": 1.550699625047855,
"choice_matches": true
},
{
"id": "duplex_ack_01",
"expected": "discard",
"choice": "enqueue",
"probabilities": [
0.04723662510514259,
0.4310700595378876,
0.5216932892799377
],
"seconds": 0.08203183300793171,
"tokens": 226,
"reference_probabilities": [
0.04766111075878143,
0.4406282901763916,
0.5117105841636658
],
"reference_choice": "enqueue",
"max_probability_error": 0.009982705116271973,
"reference_seconds": 1.50974775000941,
"choice_matches": true
},
{
"id": "duplex_ack_02",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.020713020116090775,
0.894315779209137,
0.08497121185064316
],
"seconds": 0.09132337488699704,
"tokens": 228,
"reference_probabilities": [
0.021603770554065704,
0.885711669921875,
0.0926845446228981
],
"reference_choice": "discard",
"max_probability_error": 0.008604109287261963,
"reference_seconds": 1.893160832929425,
"choice_matches": true
},
{
"id": "duplex_ack_03",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.05350596830248833,
0.5620473027229309,
0.3844466507434845
],
"seconds": 0.08070604200474918,
"tokens": 227,
"reference_probabilities": [
0.05371713265776634,
0.5573090314865112,
0.38897377252578735
],
"reference_choice": "discard",
"max_probability_error": 0.004738271236419678,
"reference_seconds": 2.3599068339681253,
"choice_matches": true
},
{
"id": "duplex_ack_04",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.06928054243326187,
0.686016321182251,
0.24470311403274536
],
"seconds": 0.07941279199440032,
"tokens": 226,
"reference_probabilities": [
0.06938163936138153,
0.6838774681091309,
0.2467409074306488
],
"reference_choice": "discard",
"max_probability_error": 0.002138853073120117,
"reference_seconds": 1.9235337499994785,
"choice_matches": true
},
{
"id": "duplex_continue_01",
"expected": "discard",
"choice": "enqueue",
"probabilities": [
0.03238857537508011,
0.14694948494434357,
0.8206619620323181
],
"seconds": 0.08031204203143716,
"tokens": 228,
"reference_probabilities": [
0.03300423175096512,
0.14912822842597961,
0.8178675174713135
],
"reference_choice": "enqueue",
"max_probability_error": 0.0027944445610046387,
"reference_seconds": 1.9747360420878977,
"choice_matches": true
},
{
"id": "duplex_negation_01",
"expected": "discard",
"choice": "enqueue",
"probabilities": [
0.23833583295345306,
0.3484111726284027,
0.41325291991233826
],
"seconds": 0.0811872499762103,
"tokens": 230,
"reference_probabilities": [
0.24283909797668457,
0.35499170422554016,
0.4021691381931305
],
"reference_choice": "enqueue",
"max_probability_error": 0.011083781719207764,
"reference_seconds": 2.1382391669321805,
"choice_matches": true
},
{
"id": "duplex_negation_02",
"expected": "discard",
"choice": "interrupt",
"probabilities": [
0.6158183217048645,
0.25162509083747864,
0.13255663216114044
],
"seconds": 0.07968854194041342,
"tokens": 228,
"reference_probabilities": [
0.6141000390052795,
0.25455474853515625,
0.13134513795375824
],
"reference_choice": "interrupt",
"max_probability_error": 0.0029296576976776123,
"reference_seconds": 1.7370499999960884,
"choice_matches": true
},
{
"id": "duplex_noise_01",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.14316071569919586,
0.7374182343482971,
0.11942102760076523
],
"seconds": 0.08046829199884087,
"tokens": 228,
"reference_probabilities": [
0.1426759511232376,
0.7348420023918152,
0.1224820464849472
],
"reference_choice": "discard",
"max_probability_error": 0.0030610188841819763,
"reference_seconds": 1.7179020419716835,
"choice_matches": true
},
{
"id": "duplex_noise_02",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.06295707821846008,
0.7446295619010925,
0.1924133449792862
],
"seconds": 0.07624604203738272,
"tokens": 224,
"reference_probabilities": [
0.06164718046784401,
0.7473939061164856,
0.1909588873386383
],
"reference_choice": "discard",
"max_probability_error": 0.0027643442153930664,
"reference_seconds": 1.8719591250410303,
"choice_matches": true
},
{
"id": "duplex_echo_01",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.03901330381631851,
0.7296719551086426,
0.2313147783279419
],
"seconds": 0.08117908309213817,
"tokens": 230,
"reference_probabilities": [
0.03965885192155838,
0.7302566170692444,
0.23008453845977783
],
"reference_choice": "discard",
"max_probability_error": 0.0012302398681640625,
"reference_seconds": 1.823017792077735,
"choice_matches": true
},
{
"id": "duplex_other_01",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.15541087090969086,
0.546040415763855,
0.29854869842529297
],
"seconds": 0.07968662504572421,
"tokens": 231,
"reference_probabilities": [
0.15892033278942108,
0.5478104948997498,
0.29326915740966797
],
"reference_choice": "discard",
"max_probability_error": 0.005279541015625,
"reference_seconds": 1.8408321660244837,
"choice_matches": true
},
{
"id": "duplex_queue_01",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.06113162636756897,
0.19676589965820312,
0.7421025037765503
],
"seconds": 0.07658904197160155,
"tokens": 231,
"reference_probabilities": [
0.06081786006689072,
0.19325658679008484,
0.7459254860877991
],
"reference_choice": "enqueue",
"max_probability_error": 0.0038229823112487793,
"reference_seconds": 1.5837899580365047,
"choice_matches": true
},
{
"id": "duplex_queue_02",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.05522550642490387,
0.13346858322620392,
0.8113059401512146
],
"seconds": 0.08071512496098876,
"tokens": 232,
"reference_probabilities": [
0.05363086238503456,
0.13678497076034546,
0.8095841407775879
],
"reference_choice": "enqueue",
"max_probability_error": 0.0033163875341415405,
"reference_seconds": 1.6148267081007361,
"choice_matches": true
},
{
"id": "duplex_queue_03",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.07100681215524673,
0.20129087567329407,
0.7277023196220398
],
"seconds": 0.08109366591088474,
"tokens": 231,
"reference_probabilities": [
0.07254466414451599,
0.20343035459518433,
0.7240249514579773
],
"reference_choice": "enqueue",
"max_probability_error": 0.0036773681640625,
"reference_seconds": 1.6989968330599368,
"choice_matches": true
},
{
"id": "duplex_queue_04",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.09818831831216812,
0.28104570508003235,
0.6207659840583801
],
"seconds": 0.08061987499240786,
"tokens": 230,
"reference_probabilities": [
0.09787087887525558,
0.28163066506385803,
0.6204984188079834
],
"reference_choice": "enqueue",
"max_probability_error": 0.0005849599838256836,
"reference_seconds": 1.7161132919136435,
"choice_matches": true
},
{
"id": "duplex_queue_05",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.1275750994682312,
0.37047719955444336,
0.5019477009773254
],
"seconds": 0.08042445802129805,
"tokens": 232,
"reference_probabilities": [
0.12692749500274658,
0.3726767897605896,
0.500395655632019
],
"reference_choice": "enqueue",
"max_probability_error": 0.0021995902061462402,
"reference_seconds": 1.7606275420403108,
"choice_matches": true
},
{
"id": "duplex_queue_06",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.042777255177497864,
0.38734444975852966,
0.5698782801628113
],
"seconds": 0.07603337499313056,
"tokens": 232,
"reference_probabilities": [
0.04404682293534279,
0.3879554271697998,
0.5679977536201477
],
"reference_choice": "enqueue",
"max_probability_error": 0.0018805265426635742,
"reference_seconds": 1.6671554159838706,
"choice_matches": true
},
{
"id": "duplex_queue_07",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.036512259393930435,
0.0908428356051445,
0.8726449012756348
],
"seconds": 0.08036591706331819,
"tokens": 228,
"reference_probabilities": [
0.03745664656162262,
0.09288956224918365,
0.8696537613868713
],
"reference_choice": "enqueue",
"max_probability_error": 0.0029911398887634277,
"reference_seconds": 1.6044136671116576,
"choice_matches": true
},
{
"id": "duplex_idle_01",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.0399501696228981,
0.30694684386253357,
0.6531029939651489
],
"seconds": 0.07956312492024153,
"tokens": 230,
"reference_probabilities": [
0.03898593783378601,
0.3061915338039398,
0.6548224687576294
],
"reference_choice": "enqueue",
"max_probability_error": 0.0017194747924804688,
"reference_seconds": 1.544517666916363,
"choice_matches": true
},
{
"id": "duplex_idle_02",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.04318610951304436,
0.158489391207695,
0.7983245253562927
],
"seconds": 0.07597420900128782,
"tokens": 224,
"reference_probabilities": [
0.04198061302304268,
0.15717433393001556,
0.8008450865745544
],
"reference_choice": "enqueue",
"max_probability_error": 0.0025205612182617188,
"reference_seconds": 1.60286049998831,
"choice_matches": true
},
{
"id": "duplex_idle_03",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.02974857948720455,
0.11033587902784348,
0.8599154949188232
],
"seconds": 0.07996929192449898,
"tokens": 226,
"reference_probabilities": [
0.029331024736166,
0.11014652252197266,
0.860522449016571
],
"reference_choice": "enqueue",
"max_probability_error": 0.0006069540977478027,
"reference_seconds": 1.6392767919460312,
"choice_matches": true
},
{
"id": "duplex_idle_04",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.0368136391043663,
0.3497421443462372,
0.6134442090988159
],
"seconds": 0.07963024999480695,
"tokens": 227,
"reference_probabilities": [
0.03727850317955017,
0.35004138946533203,
0.6126800775527954
],
"reference_choice": "enqueue",
"max_probability_error": 0.0007641315460205078,
"reference_seconds": 1.5784549170639366,
"choice_matches": true
},
{
"id": "duplex_idle_05",
"expected": "enqueue",
"choice": "discard",
"probabilities": [
0.07173047214746475,
0.4679584801197052,
0.46031105518341064
],
"seconds": 0.0805491249775514,
"tokens": 227,
"reference_probabilities": [
0.06846742331981659,
0.4646383821964264,
0.4668942093849182
],
"reference_choice": "enqueue",
"max_probability_error": 0.006583154201507568,
"reference_seconds": 1.5111134169856086,
"choice_matches": false
},
{
"id": "duplex_idle_06",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.08198647201061249,
0.1435299515724182,
0.7744835615158081
],
"seconds": 0.08135799993760884,
"tokens": 228,
"reference_probabilities": [
0.084021657705307,
0.14735808968544006,
0.7686202526092529
],
"reference_choice": "enqueue",
"max_probability_error": 0.005863308906555176,
"reference_seconds": 1.5169537500478327,
"choice_matches": true
},
{
"id": "duplex_idle_07",
"expected": "discard",
"choice": "enqueue",
"probabilities": [
0.3057652711868286,
0.33856794238090515,
0.35566675662994385
],
"seconds": 0.07646591705270112,
"tokens": 231,
"reference_probabilities": [
0.30615246295928955,
0.3364321291446686,
0.35741540789604187
],
"reference_choice": "enqueue",
"max_probability_error": 0.0021358132362365723,
"reference_seconds": 1.5602811670396477,
"choice_matches": true
},
{
"id": "duplex_idle_08",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.08129880577325821,
0.628534734249115,
0.290166437625885
],
"seconds": 0.08163958298973739,
"tokens": 227,
"reference_probabilities": [
0.08163823932409286,
0.6236112117767334,
0.29475054144859314
],
"reference_choice": "discard",
"max_probability_error": 0.004923522472381592,
"reference_seconds": 1.9073828749824315,
"choice_matches": true
},
{
"id": "duplex_answer_01",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.03204804286360741,
0.13417251408100128,
0.8337793946266174
],
"seconds": 0.08044004091061652,
"tokens": 228,
"reference_probabilities": [
0.03163864091038704,
0.13461968302726746,
0.8337417244911194
],
"reference_choice": "enqueue",
"max_probability_error": 0.0004471689462661743,
"reference_seconds": 2.5085449579637498,
"choice_matches": true
},
{
"id": "duplex_answer_02",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.03557829186320305,
0.16976748406887054,
0.7946541905403137
],
"seconds": 0.0781737919896841,
"tokens": 232,
"reference_probabilities": [
0.036002352833747864,
0.1705816388130188,
0.7934160232543945
],
"reference_choice": "enqueue",
"max_probability_error": 0.0012381672859191895,
"reference_seconds": 2.704560791957192,
"choice_matches": true
},
{
"id": "duplex_answer_03",
"expected": "enqueue",
"choice": "discard",
"probabilities": [
0.07960195094347,
0.5172246694564819,
0.40317344665527344
],
"seconds": 0.08178474998567253,
"tokens": 229,
"reference_probabilities": [
0.08029048144817352,
0.5036346912384033,
0.41607481241226196
],
"reference_choice": "discard",
"max_probability_error": 0.013589978218078613,
"reference_seconds": 2.461396458093077,
"choice_matches": true
},
{
"id": "duplex_question_stop",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.3803572356700897,
0.1342829465866089,
0.4853598475456238
],
"seconds": 0.08033037499990314,
"tokens": 232,
"reference_probabilities": [
0.3861762285232544,
0.13242685794830322,
0.48139697313308716
],
"reference_choice": "enqueue",
"max_probability_error": 0.005818992853164673,
"reference_seconds": 2.1743874580133706,
"choice_matches": true
},
{
"id": "duplex_quoted_stop",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.043692104518413544,
0.11582167446613312,
0.8404862284660339
],
"seconds": 0.08421558304689825,
"tokens": 234,
"reference_probabilities": [
0.04419822245836258,
0.11425076425075531,
0.8415510654449463
],
"reference_choice": "enqueue",
"max_probability_error": 0.0015709102153778076,
"reference_seconds": 1.9867438749643043,
"choice_matches": true
},
{
"id": "duplex_mid_ack_request",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.03667035326361656,
0.10981252789497375,
0.8535171151161194
],
"seconds": 0.08180612500291318,
"tokens": 232,
"reference_probabilities": [
0.03602280467748642,
0.10981890559196472,
0.8541582822799683
],
"reference_choice": "enqueue",
"max_probability_error": 0.0006475485861301422,
"reference_seconds": 2.4255083330208436,
"choice_matches": true
},
{
"id": "duplex_immediate_01",
"expected": "interrupt",
"choice": "discard",
"probabilities": [
0.18807342648506165,
0.5113693475723267,
0.3005572557449341
],
"seconds": 0.08018233301118016,
"tokens": 231,
"reference_probabilities": [
0.1886214166879654,
0.5195044875144958,
0.29187411069869995
],
"reference_choice": "discard",
"max_probability_error": 0.00868314504623413,
"reference_seconds": 1.987037666956894,
"choice_matches": true
},
{
"id": "duplex_wait_01",
"expected": "interrupt",
"choice": "discard",
"probabilities": [
0.17067477107048035,
0.495370090007782,
0.33395519852638245
],
"seconds": 0.08099654200486839,
"tokens": 231,
"reference_probabilities": [
0.1698099672794342,
0.49879851937294006,
0.3313915431499481
],
"reference_choice": "discard",
"max_probability_error": 0.003428429365158081,
"reference_seconds": 3.299367874977179,
"choice_matches": true
},
{
"id": "duplex_wait_02",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.16579405963420868,
0.40518027544021606,
0.42902567982673645
],
"seconds": 0.08069149998482317,
"tokens": 232,
"reference_probabilities": [
0.16112957894802094,
0.40239909291267395,
0.4364713430404663
],
"reference_choice": "enqueue",
"max_probability_error": 0.007445663213729858,
"reference_seconds": 3.0228463750099763,
"choice_matches": true
},
{
"id": "duplex_blank_01",
"expected": "discard",
"choice": "discard",
"probabilities": [
0.17456237971782684,
0.5632202625274658,
0.26221734285354614
],
"seconds": 0.08505883300676942,
"tokens": 223,
"reference_probabilities": [
0.1785687953233719,
0.558518648147583,
0.2629125416278839
],
"reference_choice": "discard",
"max_probability_error": 0.0047016143798828125,
"reference_seconds": 7.17611683299765,
"choice_matches": true
},
{
"id": "duplex_idle_correction",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.07158209383487701,
0.14543120563030243,
0.7829867005348206
],
"seconds": 0.0851495829410851,
"tokens": 232,
"reference_probabilities": [
0.07272444665431976,
0.14925770461559296,
0.7780178785324097
],
"reference_choice": "enqueue",
"max_probability_error": 0.004968822002410889,
"reference_seconds": 5.598487625014968,
"choice_matches": true
},
{
"id": "duplex_queue_constraint",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.06015212833881378,
0.14546778798103333,
0.7943800687789917
],
"seconds": 0.08386312506627291,
"tokens": 232,
"reference_probabilities": [
0.05971384048461914,
0.14688563346862793,
0.7934005260467529
],
"reference_choice": "enqueue",
"max_probability_error": 0.0014178454875946045,
"reference_seconds": 3.1079087499529123,
"choice_matches": true
},
{
"id": "duplex_late_request",
"expected": "enqueue",
"choice": "enqueue",
"probabilities": [
0.1653900444507599,
0.28522974252700806,
0.5493802428245544
],
"seconds": 0.0879432080546394,
"tokens": 234,
"reference_probabilities": [
0.16601118445396423,
0.2892772853374481,
0.5447115302085876
],
"reference_choice": "enqueue",
"max_probability_error": 0.004668712615966797,
"reference_seconds": 1.6651257500052452,
"choice_matches": true
}
],
"choices_matched": 47,
"cases": 48,
"max_probability_error": 0.013589978218078613,
"reference_business_correct": 36
}
}