d3-edge-CoreAI / comparison.json
devilyouwei's picture
Add locally converted text-only D3 Core AI model and full-duplex validation
0690c04 verified
Raw History Blame Contribute Delete
51.6 kB
[
{
"id": "en_billing",
"modality": "text",
"type": "choice",
"keys": [
"billing",
"shipping",
"account"
],
"tokens": 112,
"probabilities": [
0.9967046991738012,
0.003173908516086345,
0.00012139231011244379
],
"choice": "billing",
"end_to_end_seconds": 4.704778583953157,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.9968185424804688,
0.0030610025860369205,
0.0001203358406201005
],
"reference_choice": "billing",
"max_probability_error": 0.00011384330666752707,
"reference_seconds": 0.9592406250303611,
"choice_matches": true
},
{
"id": "en_receipt_yes",
"modality": "text",
"type": "noul",
"keys": [
"false",
"true"
],
"tokens": 97,
"probabilities": [
0.0009350354157165212,
0.9990649645842834
],
"choice": "true",
"end_to_end_seconds": 0.06896974996197969,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.0009151384583674371,
0.9990848302841187
],
"reference_choice": "true",
"max_probability_error": 1.9896957349084055e-05,
"reference_seconds": 0.9562299590324983,
"choice_matches": true
},
{
"id": "en_receipt_no",
"modality": "text",
"type": "noul",
"keys": [
"false",
"true"
],
"tokens": 101,
"probabilities": [
0.9994281199906329,
0.0005718800093671759
],
"choice": "false",
"end_to_end_seconds": 0.061143915983848274,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.9994015693664551,
0.0005984140443615615
],
"reference_choice": "false",
"max_probability_error": 2.6550624177779092e-05,
"reference_seconds": 0.8957640829030424,
"choice_matches": true
},
{
"id": "en_severity",
"modality": "text",
"type": "score",
"keys": [
"0",
"1",
"2",
"3"
],
"tokens": 102,
"probabilities": [
0.023247663416386233,
0.014125767840148172,
0.04826004530456582,
0.9143665234388998
],
"choice": "3",
"end_to_end_seconds": 0.06384520803112537,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.02332679182291031,
0.013707987032830715,
0.047914162278175354,
0.9150510430335999
],
"reference_choice": "3",
"max_probability_error": 0.0006845195947000615,
"reference_seconds": 0.8367472919635475,
"choice_matches": true
},
{
"id": "es_account",
"modality": "text",
"type": "choice",
"keys": [
"cuenta",
"pago",
"envío"
],
"tokens": 106,
"probabilities": [
0.8206464988581229,
0.0013023086409113583,
0.17805119250096557
],
"choice": "cuenta",
"end_to_end_seconds": 0.06666487501934171,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.8160568475723267,
0.0012253460008651018,
0.18271781504154205
],
"reference_choice": "cuenta",
"max_probability_error": 0.004666622540576482,
"reference_seconds": 0.837753374944441,
"choice_matches": true
},
{
"id": "zh_photo",
"modality": "text",
"type": "choice",
"keys": [
"摄影",
"写作",
"闲聊"
],
"tokens": 117,
"probabilities": [
0.9998393001136836,
3.488917131004768e-05,
0.00012581071500638612
],
"choice": "摄影",
"end_to_end_seconds": 0.07095762505196035,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.9998327493667603,
3.669255238492042e-05,
0.00013051324640400708
],
"reference_choice": "摄影",
"max_probability_error": 6.550746923350559e-06,
"reference_seconds": 0.9610827500000596,
"choice_matches": true
},
{
"id": "zh_write",
"modality": "text",
"type": "choice",
"keys": [
"摄影",
"写作",
"闲聊"
],
"tokens": 117,
"probabilities": [
0.0348665825414572,
0.9633777973855959,
0.0017556200729468056
],
"choice": "写作",
"end_to_end_seconds": 0.06214854202698916,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.03644991293549538,
0.9617894291877747,
0.0017606894252821803
],
"reference_choice": "写作",
"max_probability_error": 0.001588368197821266,
"reference_seconds": 1.1819746660767123,
"choice_matches": true
},
{
"id": "zh_complete",
"modality": "text",
"type": "noul",
"keys": [
"false",
"true"
],
"tokens": 100,
"probabilities": [
0.05187394207649316,
0.9481260579235069
],
"choice": "true",
"end_to_end_seconds": 0.06265779200475663,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.05357705056667328,
0.9464228749275208
],
"reference_choice": "true",
"max_probability_error": 0.0017031829959861522,
"reference_seconds": 0.9138488749740645,
"choice_matches": true
},
{
"id": "zh_incomplete",
"modality": "text",
"type": "noul",
"keys": [
"false",
"true"
],
"tokens": 96,
"probabilities": [
0.9873243070746749,
0.012675692925325145
],
"choice": "false",
"end_to_end_seconds": 0.0630836250493303,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.9866238236427307,
0.013376150280237198
],
"reference_choice": "false",
"max_probability_error": 0.0007004834319441766,
"reference_seconds": 0.8583499160595238,
"choice_matches": true
},
{
"id": "zh_failure",
"modality": "text",
"type": "choice",
"keys": [
"成功",
"失败",
"等待"
],
"tokens": 115,
"probabilities": [
0.0005653045553556631,
0.9603952327527144,
0.03903946269193009
],
"choice": "失败",
"end_to_end_seconds": 0.06020887498743832,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.0005506259622052312,
0.9596866965293884,
0.03976264223456383
],
"reference_choice": "失败",
"max_probability_error": 0.0007231795426337409,
"reference_seconds": 0.9902184169040993,
"choice_matches": true
},
{
"id": "zh_urgency",
"modality": "text",
"type": "score",
"keys": [
"0",
"1",
"2",
"3"
],
"tokens": 108,
"probabilities": [
0.9797662450595488,
0.01630116107996746,
0.002195700708535573,
0.0017368931519482734
],
"choice": "0",
"end_to_end_seconds": 0.058163207955658436,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.9800215363502502,
0.016013935208320618,
0.0021757411304861307,
0.0017887389985844493
],
"reference_choice": "0",
"max_probability_error": 0.00028722587164684357,
"reference_seconds": 0.8716891669901088,
"choice_matches": true
},
{
"id": "zh_ambiguous",
"modality": "text",
"type": "choice",
"keys": [
"询问",
"拍照",
"写作"
],
"tokens": 117,
"probabilities": [
0.9873235945095953,
0.0013211292784958406,
0.01135527621190901
],
"choice": "询问",
"end_to_end_seconds": 0.06174908403772861,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.9874310493469238,
0.0012258542701601982,
0.011343155056238174
],
"reference_choice": "询问",
"max_probability_error": 0.00010745483732854666,
"reference_seconds": 0.8878887500613928,
"choice_matches": true
},
{
"id": "zh_negation",
"modality": "text",
"type": "noul",
"keys": [
"false",
"true"
],
"tokens": 98,
"probabilities": [
0.9730555471210252,
0.026944452878974963
],
"choice": "false",
"end_to_end_seconds": 0.059276791056618094,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.9733514189720154,
0.026648661121726036
],
"reference_choice": "false",
"max_probability_error": 0.00029587185099022584,
"reference_seconds": 0.8123413749272004,
"choice_matches": true
},
{
"id": "zh_json",
"modality": "text",
"type": "choice",
"keys": [
"成功",
"失败",
"等待"
],
"tokens": 111,
"probabilities": [
6.080650784976944e-05,
0.9985972973829926,
0.00134189610915757
],
"choice": "失败",
"end_to_end_seconds": 0.05723312497138977,
"expected": null,
"suite": "general",
"reference_probabilities": [
5.890630200156011e-05,
0.998606264591217,
0.0013348645297810435
],
"reference_choice": "失败",
"max_probability_error": 8.967208224408552e-06,
"reference_seconds": 0.9132232909323648,
"choice_matches": true
},
{
"id": "en_ten_options",
"modality": "text",
"type": "choice",
"keys": [
"0",
"1",
"2",
"3",
"4",
"5",
"6",
"7",
"8",
"9"
],
"tokens": 152,
"probabilities": [
0.004225672394648097,
0.0011894069653672075,
0.971975838323805,
5.90924876919151e-05,
0.0003250735662990723,
0.00011953259928416013,
0.000549238651651673,
0.006027731982460974,
0.00018880039154618676,
0.015339612637245634
],
"choice": "2",
"end_to_end_seconds": 0.06183316698297858,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.0042327879928052425,
0.0012714171316474676,
0.971185028553009,
6.0139478591736406e-05,
0.0003370257036294788,
0.00012199221237096936,
0.0005663561751134694,
0.00637219101190567,
0.0001901781652122736,
0.015662873163819313
],
"reference_choice": "2",
"max_probability_error": 0.0007908097707959305,
"reference_seconds": 1.0713317920453846,
"choice_matches": true
},
{
"id": "one_option",
"modality": "text",
"type": "choice",
"keys": [
"only"
],
"tokens": 90,
"probabilities": [
1.0
],
"choice": "only",
"end_to_end_seconds": 0.055508290999569,
"expected": null,
"suite": "general",
"reference_probabilities": [
1.0
],
"reference_choice": "only",
"max_probability_error": 0.0,
"reference_seconds": 0.7906684160698205,
"choice_matches": true
},
{
"id": "empty_state",
"modality": "text",
"type": "noul",
"keys": [
"false",
"true"
],
"tokens": 94,
"probabilities": [
0.9666735695988837,
0.03332643040111631
],
"choice": "false",
"end_to_end_seconds": 0.056953666964545846,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.966280996799469,
0.033719006925821304
],
"reference_choice": "false",
"max_probability_error": 0.0003925765247049934,
"reference_seconds": 0.8284598749596626,
"choice_matches": true
},
{
"id": "long_text",
"modality": "text",
"type": "choice",
"keys": [
"refund",
"shipping"
],
"tokens": 433,
"probabilities": [
0.9992611891371558,
0.0007388108628442402
],
"choice": "refund",
"end_to_end_seconds": 0.05793479201383889,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.9992407560348511,
0.000759303686209023
],
"reference_choice": "refund",
"max_probability_error": 2.0492823364782814e-05,
"reference_seconds": 2.7816366250626743,
"choice_matches": true
},
{
"id": "en_billing_reversed",
"modality": "text",
"type": "choice",
"keys": [
"account",
"shipping",
"billing"
],
"tokens": 112,
"probabilities": [
0.00017229889995540166,
0.005839408458566054,
0.9939882926414785
],
"choice": "billing",
"end_to_end_seconds": 0.06136983307078481,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.00017491288599558175,
0.005699815694242716,
0.994125247001648
],
"reference_choice": "billing",
"max_probability_error": 0.00013959276432333814,
"reference_seconds": 0.9026339160045609,
"choice_matches": true
},
{
"id": "es_account_reversed",
"modality": "text",
"type": "choice",
"keys": [
"envío",
"pago",
"cuenta"
],
"tokens": 106,
"probabilities": [
0.15988822967617503,
0.0014988406229400862,
0.8386129297008847
],
"choice": "cuenta",
"end_to_end_seconds": 0.07257750001735985,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.16495831310749054,
0.001530324574559927,
0.8335114121437073
],
"reference_choice": "cuenta",
"max_probability_error": 0.005101517557177471,
"reference_seconds": 0.97938629204873,
"choice_matches": true
},
{
"id": "zh_photo_reversed",
"modality": "text",
"type": "choice",
"keys": [
"闲聊",
"写作",
"摄影"
],
"tokens": 117,
"probabilities": [
0.00023131879038623356,
4.508569764270688e-05,
0.9997235955119711
],
"choice": "摄影",
"end_to_end_seconds": 0.061316207982599735,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.00023583779693581164,
4.526936390902847e-05,
0.9997188448905945
],
"reference_choice": "摄影",
"max_probability_error": 4.750621376614639e-06,
"reference_seconds": 1.1876654999796301,
"choice_matches": true
},
{
"id": "zh_write_reversed",
"modality": "text",
"type": "choice",
"keys": [
"闲聊",
"写作",
"摄影"
],
"tokens": 117,
"probabilities": [
0.001298186924197409,
0.9889334119394662,
0.009768401136336279
],
"choice": "写作",
"end_to_end_seconds": 0.06560208299197257,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.0012285992270335555,
0.9890910983085632,
0.009680334478616714
],
"reference_choice": "写作",
"max_probability_error": 0.00015768636909707645,
"reference_seconds": 0.9455100829945877,
"choice_matches": true
},
{
"id": "zh_failure_reversed",
"modality": "text",
"type": "choice",
"keys": [
"等待",
"失败",
"成功"
],
"tokens": 115,
"probabilities": [
0.0377791663012688,
0.9614243578445482,
0.000796475854183018
],
"choice": "失败",
"end_to_end_seconds": 0.06017183302901685,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.0394730269908905,
0.9597182273864746,
0.000808742712251842
],
"reference_choice": "失败",
"max_probability_error": 0.0017061304580735515,
"reference_seconds": 0.9902247909922153,
"choice_matches": true
},
{
"id": "zh_ambiguous_reversed",
"modality": "text",
"type": "choice",
"keys": [
"写作",
"拍照",
"询问"
],
"tokens": 117,
"probabilities": [
0.012942807992159527,
0.001024783462247127,
0.9860324085455934
],
"choice": "询问",
"end_to_end_seconds": 0.05979891598690301,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.013815107755362988,
0.0010390413226559758,
0.9851458072662354
],
"reference_choice": "询问",
"max_probability_error": 0.0008866012793580236,
"reference_seconds": 0.924791333032772,
"choice_matches": true
},
{
"id": "zh_json_reversed",
"modality": "text",
"type": "choice",
"keys": [
"等待",
"失败",
"成功"
],
"tokens": 111,
"probabilities": [
0.0010409366689280572,
0.9988988429175736,
6.0220413498355025e-05
],
"choice": "失败",
"end_to_end_seconds": 0.061546499957330525,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.0010691973147913814,
0.9988671541213989,
6.365815352182835e-05
],
"reference_choice": "失败",
"max_probability_error": 3.168879617465592e-05,
"reference_seconds": 0.8861740420106798,
"choice_matches": true
},
{
"id": "en_ten_options_reversed",
"modality": "text",
"type": "choice",
"keys": [
"9",
"8",
"7",
"6",
"5",
"4",
"3",
"2",
"1",
"0"
],
"tokens": 152,
"probabilities": [
0.013898673441840799,
0.00022282842811656705,
0.005271534350530006,
0.0005357954604744423,
0.0001539025938945331,
0.0001586649674127105,
2.725876234088219e-05,
0.9726926979974864,
0.0041623443396139785,
0.0028762996582898767
],
"choice": "2",
"end_to_end_seconds": 0.06301770906429738,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.013970260508358479,
0.00022368931968230754,
0.005643855314701796,
0.0005588575731962919,
0.0001519664074294269,
0.00016414382844232023,
2.832768041116651e-05,
0.9725043773651123,
0.003967889118939638,
0.002786590252071619
],
"reference_choice": "2",
"max_probability_error": 0.0003723209641717899,
"reference_seconds": 1.1101172079797834,
"choice_matches": true
},
{
"id": "long_text_reversed",
"modality": "text",
"type": "choice",
"keys": [
"shipping",
"refund"
],
"tokens": 433,
"probabilities": [
0.0004540697084874103,
0.9995459302915126
],
"choice": "refund",
"end_to_end_seconds": 0.06365333299618214,
"expected": null,
"suite": "general",
"reference_probabilities": [
0.00044810003601014614,
0.9995518326759338
],
"reference_choice": "refund",
"max_probability_error": 5.969672477264178e-06,
"reference_seconds": 2.81223541602958,
"choice_matches": true
},
{
"id": "duplex_stop_01",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 320,
"probabilities": [
0.36925801930919255,
0.28427324736144605,
0.34646873332936146
],
"choice": "interrupt",
"end_to_end_seconds": 0.06667500000912696,
"expected": "interrupt",
"suite": "full_duplex",
"reference_probabilities": [
0.3725797235965729,
0.2848637104034424,
0.34255659580230713
],
"reference_choice": "interrupt",
"max_probability_error": 0.003912137527054327,
"reference_seconds": 2.154552042018622,
"choice_matches": true
},
{
"id": "duplex_stop_02",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 320,
"probabilities": [
0.09961632678900385,
0.4228922080417295,
0.47749146516926666
],
"choice": "enqueue",
"end_to_end_seconds": 0.06271600001491606,
"expected": "interrupt",
"suite": "full_duplex",
"reference_probabilities": [
0.0985235646367073,
0.4244298040866852,
0.4770466089248657
],
"reference_choice": "enqueue",
"max_probability_error": 0.0015375960449556803,
"reference_seconds": 2.044221166986972,
"choice_matches": true
},
{
"id": "duplex_stop_03",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 321,
"probabilities": [
0.4476812623612501,
0.4019674639023257,
0.1503512737364243
],
"choice": "interrupt",
"end_to_end_seconds": 0.061337166000157595,
"expected": "interrupt",
"suite": "full_duplex",
"reference_probabilities": [
0.43477410078048706,
0.4139057993888855,
0.15132005512714386
],
"reference_choice": "interrupt",
"max_probability_error": 0.012907161580763049,
"reference_seconds": 2.2079034170601517,
"choice_matches": true
},
{
"id": "duplex_stop_04",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 322,
"probabilities": [
0.3121490077614103,
0.29746391695469465,
0.390387075283895
],
"choice": "enqueue",
"end_to_end_seconds": 0.06084333301987499,
"expected": "interrupt",
"suite": "full_duplex",
"reference_probabilities": [
0.3236846327781677,
0.2929101288318634,
0.38340529799461365
],
"reference_choice": "enqueue",
"max_probability_error": 0.011535625016757423,
"reference_seconds": 2.1794147910550237,
"choice_matches": true
},
{
"id": "duplex_correct_01",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 324,
"probabilities": [
0.08999144907868228,
0.5631892421055145,
0.34681930881580314
],
"choice": "discard",
"end_to_end_seconds": 0.059515041997656226,
"expected": "interrupt",
"suite": "full_duplex",
"reference_probabilities": [
0.08759114146232605,
0.56805819272995,
0.344350665807724
],
"reference_choice": "discard",
"max_probability_error": 0.004868950624435486,
"reference_seconds": 2.1173955840058625,
"choice_matches": true
},
{
"id": "duplex_correct_02",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 324,
"probabilities": [
0.15449782794531777,
0.3732968993524262,
0.47220527270225593
],
"choice": "enqueue",
"end_to_end_seconds": 0.058978416956961155,
"expected": "interrupt",
"suite": "full_duplex",
"reference_probabilities": [
0.15848976373672485,
0.363983690738678,
0.47752657532691956
],
"reference_choice": "enqueue",
"max_probability_error": 0.009313208613748236,
"reference_seconds": 2.1236701250309125,
"choice_matches": true
},
{
"id": "duplex_correct_03",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 323,
"probabilities": [
0.14239350315981378,
0.49466656928305286,
0.36293992755713333
],
"choice": "discard",
"end_to_end_seconds": 0.06071037508081645,
"expected": "interrupt",
"suite": "full_duplex",
"reference_probabilities": [
0.13985002040863037,
0.5068005919456482,
0.35334938764572144
],
"reference_choice": "discard",
"max_probability_error": 0.012134022662595334,
"reference_seconds": 2.1158938329899684,
"choice_matches": true
},
{
"id": "duplex_switch_01",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 325,
"probabilities": [
0.10244300525216662,
0.5298673441841425,
0.3676896505636909
],
"choice": "discard",
"end_to_end_seconds": 0.06122833304107189,
"expected": "interrupt",
"suite": "full_duplex",
"reference_probabilities": [
0.10244057327508926,
0.5285248756408691,
0.3690345883369446
],
"reference_choice": "discard",
"max_probability_error": 0.0013449377732536738,
"reference_seconds": 2.1220966250402853,
"choice_matches": true
},
{
"id": "duplex_urgent_01",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 325,
"probabilities": [
0.26214412519420593,
0.489283329704586,
0.248572545101208
],
"choice": "discard",
"end_to_end_seconds": 0.06692820903845131,
"expected": "interrupt",
"suite": "full_duplex",
"reference_probabilities": [
0.25545012950897217,
0.49854594469070435,
0.24600394070148468
],
"reference_choice": "discard",
"max_probability_error": 0.009262614986118356,
"reference_seconds": 2.075061084004119,
"choice_matches": true
},
{
"id": "duplex_ack_01",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 318,
"probabilities": [
0.07963558792521354,
0.3749885085607277,
0.5453759035140588
],
"choice": "enqueue",
"end_to_end_seconds": 0.06161624996457249,
"expected": "discard",
"suite": "full_duplex",
"reference_probabilities": [
0.07771968841552734,
0.37838810682296753,
0.5438922047615051
],
"reference_choice": "enqueue",
"max_probability_error": 0.0033995982622398446,
"reference_seconds": 2.0976727079832926,
"choice_matches": true
},
{
"id": "duplex_ack_02",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 320,
"probabilities": [
0.047980765908528546,
0.4190138556523637,
0.5330053784391077
],
"choice": "enqueue",
"end_to_end_seconds": 0.0604598339414224,
"expected": "discard",
"suite": "full_duplex",
"reference_probabilities": [
0.048545438796281815,
0.4277784526348114,
0.5236760973930359
],
"reference_choice": "enqueue",
"max_probability_error": 0.00932928104607178,
"reference_seconds": 2.1040617909748107,
"choice_matches": true
},
{
"id": "duplex_ack_03",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 319,
"probabilities": [
0.061341726904610365,
0.411213218453489,
0.5274450546419007
],
"choice": "enqueue",
"end_to_end_seconds": 0.060387250036001205,
"expected": "discard",
"suite": "full_duplex",
"reference_probabilities": [
0.06230482831597328,
0.3923381268978119,
0.5453571081161499
],
"reference_choice": "enqueue",
"max_probability_error": 0.018875091555677126,
"reference_seconds": 2.1018672079080716,
"choice_matches": true
},
{
"id": "duplex_ack_04",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 318,
"probabilities": [
0.03816007622578243,
0.6283586562906518,
0.3334812674835658
],
"choice": "discard",
"end_to_end_seconds": 0.059120125020854175,
"expected": "discard",
"suite": "full_duplex",
"reference_probabilities": [
0.03831863775849342,
0.6333273649215698,
0.32835397124290466
],
"reference_choice": "discard",
"max_probability_error": 0.005127296240661161,
"reference_seconds": 2.0880519170314074,
"choice_matches": true
},
{
"id": "duplex_continue_01",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 320,
"probabilities": [
0.032283539276670464,
0.3601747640585033,
0.6075416966648263
],
"choice": "enqueue",
"end_to_end_seconds": 0.05947733297944069,
"expected": "discard",
"suite": "full_duplex",
"reference_probabilities": [
0.03161618486046791,
0.37331023812294006,
0.5950735807418823
],
"reference_choice": "enqueue",
"max_probability_error": 0.01313547406443677,
"reference_seconds": 2.115650082938373,
"choice_matches": true
},
{
"id": "duplex_negation_01",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 322,
"probabilities": [
0.1273541660883302,
0.23633057889995202,
0.6363152550117177
],
"choice": "enqueue",
"end_to_end_seconds": 0.061438290984369814,
"expected": "discard",
"suite": "full_duplex",
"reference_probabilities": [
0.12637580931186676,
0.2476085126399994,
0.6260156631469727
],
"reference_choice": "enqueue",
"max_probability_error": 0.011277933740047374,
"reference_seconds": 2.603881916962564,
"choice_matches": true
},
{
"id": "duplex_negation_02",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 320,
"probabilities": [
0.15291998390093559,
0.5575064743055403,
0.28957354179352396
],
"choice": "discard",
"end_to_end_seconds": 0.06080870900768787,
"expected": "discard",
"suite": "full_duplex",
"reference_probabilities": [
0.1580505073070526,
0.5516903400421143,
0.2902591824531555
],
"reference_choice": "discard",
"max_probability_error": 0.005816134263426087,
"reference_seconds": 2.478561875061132,
"choice_matches": true
},
{
"id": "duplex_noise_01",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 321,
"probabilities": [
0.08495002511799675,
0.6141038432610582,
0.30094613162094525
],
"choice": "discard",
"end_to_end_seconds": 0.06056783301755786,
"expected": "discard",
"suite": "full_duplex",
"reference_probabilities": [
0.08534883707761765,
0.6181000471115112,
0.2965511083602905
],
"reference_choice": "discard",
"max_probability_error": 0.004395023260654718,
"reference_seconds": 2.6364804579643533,
"choice_matches": true
},
{
"id": "duplex_noise_02",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 315,
"probabilities": [
0.031172420567157504,
0.6498952576403955,
0.318932321792447
],
"choice": "discard",
"end_to_end_seconds": 0.060432375059463084,
"expected": "discard",
"suite": "full_duplex",
"reference_probabilities": [
0.030620092526078224,
0.6666684746742249,
0.3027114272117615
],
"reference_choice": "discard",
"max_probability_error": 0.01677321703382939,
"reference_seconds": 2.5116792910266668,
"choice_matches": true
},
{
"id": "duplex_echo_01",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 323,
"probabilities": [
0.030969814208492326,
0.5078372306657072,
0.4611929551258005
],
"choice": "discard",
"end_to_end_seconds": 0.05969308305066079,
"expected": "discard",
"suite": "full_duplex",
"reference_probabilities": [
0.03078923560678959,
0.5066130757331848,
0.46259763836860657
],
"reference_choice": "discard",
"max_probability_error": 0.0014046832428060707,
"reference_seconds": 2.5884792499709874,
"choice_matches": true
},
{
"id": "duplex_other_01",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 328,
"probabilities": [
0.05239836011394757,
0.729776372632972,
0.21782526725308043
],
"choice": "discard",
"end_to_end_seconds": 0.0587671670364216,
"expected": "discard",
"suite": "full_duplex",
"reference_probabilities": [
0.052182890474796295,
0.7242393493652344,
0.22357770800590515
],
"reference_choice": "discard",
"max_probability_error": 0.005752440752824717,
"reference_seconds": 2.356515208957717,
"choice_matches": true
},
{
"id": "duplex_queue_01",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 323,
"probabilities": [
0.08632894942256907,
0.40525975317069746,
0.5084112974067334
],
"choice": "enqueue",
"end_to_end_seconds": 0.059953041025437415,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.08526115119457245,
0.39749082922935486,
0.5172479748725891
],
"reference_choice": "enqueue",
"max_probability_error": 0.008836677465855702,
"reference_seconds": 2.2746908330591395,
"choice_matches": true
},
{
"id": "duplex_queue_02",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 324,
"probabilities": [
0.08665169313664238,
0.38401823876268754,
0.5293300681006701
],
"choice": "enqueue",
"end_to_end_seconds": 0.05933887499850243,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.08464257419109344,
0.3730492889881134,
0.5423081517219543
],
"reference_choice": "enqueue",
"max_probability_error": 0.012978083621284275,
"reference_seconds": 2.160552999936044,
"choice_matches": true
},
{
"id": "duplex_queue_03",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 323,
"probabilities": [
0.06492202507384029,
0.5751809164396222,
0.35989705848653747
],
"choice": "discard",
"end_to_end_seconds": 0.0598918329924345,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.06729138642549515,
0.5767751932144165,
0.35593339800834656
],
"reference_choice": "discard",
"max_probability_error": 0.003963660478190911,
"reference_seconds": 2.1083541250554845,
"choice_matches": true
},
{
"id": "duplex_queue_04",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 322,
"probabilities": [
0.0787192963509872,
0.3165707827036414,
0.6047099209453713
],
"choice": "enqueue",
"end_to_end_seconds": 0.06024570798035711,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.07793932408094406,
0.31252092123031616,
0.609539806842804
],
"reference_choice": "enqueue",
"max_probability_error": 0.0048298858974326064,
"reference_seconds": 2.5078482090029866,
"choice_matches": true
},
{
"id": "duplex_queue_05",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 324,
"probabilities": [
0.07125473397953433,
0.766075926756357,
0.16266933926410856
],
"choice": "discard",
"end_to_end_seconds": 0.058723207912407815,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.07154783606529236,
0.763809323310852,
0.1646428406238556
],
"reference_choice": "discard",
"max_probability_error": 0.002266603445505,
"reference_seconds": 2.687678375048563,
"choice_matches": true
},
{
"id": "duplex_queue_06",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 324,
"probabilities": [
0.03492957278245516,
0.707586016074946,
0.25748441114259885
],
"choice": "discard",
"end_to_end_seconds": 0.059189834049902856,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.036303821951150894,
0.6983221769332886,
0.26537397503852844
],
"reference_choice": "discard",
"max_probability_error": 0.009263839141657404,
"reference_seconds": 2.302886249963194,
"choice_matches": true
},
{
"id": "duplex_queue_07",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 320,
"probabilities": [
0.07105439547001287,
0.3090688295102327,
0.6198767750197545
],
"choice": "enqueue",
"end_to_end_seconds": 0.059386457898654044,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.07182841002941132,
0.31347477436065674,
0.6146968603134155
],
"reference_choice": "enqueue",
"max_probability_error": 0.0051799147063389395,
"reference_seconds": 2.5398947920184582,
"choice_matches": true
},
{
"id": "duplex_idle_01",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 321,
"probabilities": [
0.043018641754313455,
0.5924647905462969,
0.36451656769938956
],
"choice": "discard",
"end_to_end_seconds": 0.05938079103361815,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.043696679174900055,
0.5834992527961731,
0.37280407547950745
],
"reference_choice": "discard",
"max_probability_error": 0.008965537750123764,
"reference_seconds": 2.2472779999952763,
"choice_matches": true
},
{
"id": "duplex_idle_02",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 314,
"probabilities": [
0.028384471265278037,
0.7882387947233418,
0.18337673401138
],
"choice": "discard",
"end_to_end_seconds": 0.05945125001017004,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.028607148677110672,
0.7788572311401367,
0.19253557920455933
],
"reference_choice": "discard",
"max_probability_error": 0.009381563583205077,
"reference_seconds": 2.057773832930252,
"choice_matches": true
},
{
"id": "duplex_idle_03",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 317,
"probabilities": [
0.06194274730611633,
0.3579116094954958,
0.5801456431983879
],
"choice": "enqueue",
"end_to_end_seconds": 0.058600666001439095,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.06294164061546326,
0.3607334494590759,
0.5763249397277832
],
"reference_choice": "enqueue",
"max_probability_error": 0.0038207034706047027,
"reference_seconds": 2.1569439170416445,
"choice_matches": true
},
{
"id": "duplex_idle_04",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 318,
"probabilities": [
0.04926003054959192,
0.516774661050348,
0.4339653084000602
],
"choice": "discard",
"end_to_end_seconds": 0.060015207971446216,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.0507386140525341,
0.5229406952857971,
0.4263206124305725
],
"reference_choice": "discard",
"max_probability_error": 0.0076446959694876915,
"reference_seconds": 2.0955650419928133,
"choice_matches": true
},
{
"id": "duplex_idle_05",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 318,
"probabilities": [
0.07047516906773532,
0.5816586686209848,
0.3478661623112799
],
"choice": "discard",
"end_to_end_seconds": 0.06003241706639528,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.07387121021747589,
0.5652998089790344,
0.36082905530929565
],
"reference_choice": "discard",
"max_probability_error": 0.016358859641950385,
"reference_seconds": 2.3158621250186116,
"choice_matches": true
},
{
"id": "duplex_idle_06",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 319,
"probabilities": [
0.054862263473344386,
0.261953842611128,
0.6831838939155277
],
"choice": "enqueue",
"end_to_end_seconds": 0.06037212489172816,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.05269906669855118,
0.2537921369075775,
0.6935088634490967
],
"reference_choice": "enqueue",
"max_probability_error": 0.010324969533568984,
"reference_seconds": 2.3967140420572832,
"choice_matches": true
},
{
"id": "duplex_idle_07",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 322,
"probabilities": [
0.21905552347367904,
0.4170784198680774,
0.36386605665824345
],
"choice": "discard",
"end_to_end_seconds": 0.06213058391585946,
"expected": "discard",
"suite": "full_duplex",
"reference_probabilities": [
0.22615335881710052,
0.40118929743766785,
0.37265732884407043
],
"reference_choice": "discard",
"max_probability_error": 0.015889122430409575,
"reference_seconds": 3.188934582984075,
"choice_matches": true
},
{
"id": "duplex_idle_08",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 318,
"probabilities": [
0.025792290663845002,
0.8074185876195806,
0.16678912171657437
],
"choice": "discard",
"end_to_end_seconds": 0.061945791938342154,
"expected": "discard",
"suite": "full_duplex",
"reference_probabilities": [
0.025497060269117355,
0.8071024417877197,
0.16740049421787262
],
"reference_choice": "discard",
"max_probability_error": 0.0006113725012982518,
"reference_seconds": 2.9485161670017987,
"choice_matches": true
},
{
"id": "duplex_answer_01",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 319,
"probabilities": [
0.04394860316058848,
0.3481044061363044,
0.6079469907031071
],
"choice": "enqueue",
"end_to_end_seconds": 0.06280475005041808,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.04391104727983475,
0.33391234278678894,
0.6221765875816345
],
"reference_choice": "enqueue",
"max_probability_error": 0.014229596878527428,
"reference_seconds": 2.417675583041273,
"choice_matches": true
},
{
"id": "duplex_answer_02",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 323,
"probabilities": [
0.04722920938722117,
0.4037745853746926,
0.5489962052380862
],
"choice": "enqueue",
"end_to_end_seconds": 0.06394199992064387,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.04911813884973526,
0.40117576718330383,
0.5497061014175415
],
"reference_choice": "enqueue",
"max_probability_error": 0.002598818191388741,
"reference_seconds": 2.5277819159673527,
"choice_matches": true
},
{
"id": "duplex_answer_03",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 320,
"probabilities": [
0.07882625495120331,
0.4897181337438413,
0.4314556113049553
],
"choice": "discard",
"end_to_end_seconds": 0.06554308300837874,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.07870376110076904,
0.4732263386249542,
0.44806990027427673
],
"reference_choice": "discard",
"max_probability_error": 0.016614288969321434,
"reference_seconds": 2.4053917500423267,
"choice_matches": true
},
{
"id": "duplex_question_stop",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 324,
"probabilities": [
0.2356445421335971,
0.4050335003776703,
0.3593219574887327
],
"choice": "discard",
"end_to_end_seconds": 0.06298854190390557,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.2321690022945404,
0.40891507267951965,
0.35891595482826233
],
"reference_choice": "discard",
"max_probability_error": 0.0038815723018493786,
"reference_seconds": 2.434441916993819,
"choice_matches": true
},
{
"id": "duplex_quoted_stop",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 326,
"probabilities": [
0.172745804964821,
0.3501416970290177,
0.47711249800616123
],
"choice": "enqueue",
"end_to_end_seconds": 0.06202849990222603,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.1802801489830017,
0.3509426414966583,
0.4687771797180176
],
"reference_choice": "enqueue",
"max_probability_error": 0.008335318288143656,
"reference_seconds": 2.412521415972151,
"choice_matches": true
},
{
"id": "duplex_mid_ack_request",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 324,
"probabilities": [
0.07656011712720605,
0.3853191846102425,
0.5381206982625515
],
"choice": "enqueue",
"end_to_end_seconds": 0.062182707944884896,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.07440771162509918,
0.39047563076019287,
0.5351166725158691
],
"reference_choice": "enqueue",
"max_probability_error": 0.005156446149950367,
"reference_seconds": 2.7441802920075133,
"choice_matches": true
},
{
"id": "duplex_immediate_01",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 323,
"probabilities": [
0.1417682995473252,
0.4817785049006149,
0.37645319555205986
],
"choice": "discard",
"end_to_end_seconds": 0.06371020805090666,
"expected": "interrupt",
"suite": "full_duplex",
"reference_probabilities": [
0.14223743975162506,
0.4795777499675751,
0.37818479537963867
],
"reference_choice": "discard",
"max_probability_error": 0.002200754933039839,
"reference_seconds": 2.577961500035599,
"choice_matches": true
},
{
"id": "duplex_wait_01",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 323,
"probabilities": [
0.06340534464139812,
0.4902856188013375,
0.44630903655726434
],
"choice": "discard",
"end_to_end_seconds": 0.06390249996911734,
"expected": "interrupt",
"suite": "full_duplex",
"reference_probabilities": [
0.061347734183073044,
0.5008657574653625,
0.4377864599227905
],
"reference_choice": "discard",
"max_probability_error": 0.010580138664025074,
"reference_seconds": 2.4937992088962346,
"choice_matches": true
},
{
"id": "duplex_wait_02",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 324,
"probabilities": [
0.05975631583036526,
0.33989515479879506,
0.6003485293708397
],
"choice": "enqueue",
"end_to_end_seconds": 0.062225042027421296,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.0596485361456871,
0.33587583899497986,
0.6044756174087524
],
"reference_choice": "enqueue",
"max_probability_error": 0.0041270880379127295,
"reference_seconds": 2.5074617080390453,
"choice_matches": true
},
{
"id": "duplex_blank_01",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 318,
"probabilities": [
0.08969787823679795,
0.7349016781047959,
0.17540044365840612
],
"choice": "discard",
"end_to_end_seconds": 0.06231374992057681,
"expected": "discard",
"suite": "full_duplex",
"reference_probabilities": [
0.08804052323102951,
0.7373180985450745,
0.1746414303779602
],
"reference_choice": "discard",
"max_probability_error": 0.002416420440278566,
"reference_seconds": 2.9314769170014188,
"choice_matches": true
},
{
"id": "duplex_idle_correction",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 323,
"probabilities": [
0.11577608680906033,
0.4743426366018017,
0.4098812765891379
],
"choice": "discard",
"end_to_end_seconds": 0.0622567500686273,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.11359476298093796,
0.4820769131183624,
0.4043283462524414
],
"reference_choice": "discard",
"max_probability_error": 0.007734276516560723,
"reference_seconds": 2.8050833749584854,
"choice_matches": true
},
{
"id": "duplex_queue_constraint",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 324,
"probabilities": [
0.08326479044087588,
0.3834820274804395,
0.5332531820786847
],
"choice": "enqueue",
"end_to_end_seconds": 0.06318725005257875,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.08571639657020569,
0.3758048713207245,
0.538478672504425
],
"reference_choice": "enqueue",
"max_probability_error": 0.007677156159715037,
"reference_seconds": 2.5240845000371337,
"choice_matches": true
},
{
"id": "duplex_late_request",
"modality": "text",
"type": "choice",
"keys": [
"interrupt",
"discard",
"enqueue"
],
"tokens": 326,
"probabilities": [
0.13812516115511783,
0.4212357585373965,
0.44063908030748555
],
"choice": "enqueue",
"end_to_end_seconds": 0.06448208296205848,
"expected": "enqueue",
"suite": "full_duplex",
"reference_probabilities": [
0.13780291378498077,
0.4278469979763031,
0.4343501031398773
],
"reference_choice": "enqueue",
"max_probability_error": 0.006611239438906624,
"reference_seconds": 3.391970333061181,
"choice_matches": true
}
]