Skip to content

feat(sob): new Interfaze scores #11

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Khurdhula-Harshavardhan merged 2 commits into from
May 22, 2026
Merged

feat(sob): new Interfaze scores #11

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
2cdea65
feat(sob): new Interfaze scores
Abhinavexist May 22, 2026
17a2800
chore: remove .DS_store
Abhinavexist May 22, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
115 changes: 115 additions & 0 deletions data/audio_responses/response_interfaze-beta_audio.jsonl
View file
Open in desktop

Large diffs are not rendered by default.

115 changes: 115 additions & 0 deletions data/evaluation/audio/interfaze-beta/eval_records.jsonl
View file
Open in desktop

Large diffs are not rendered by default.

140 changes: 70 additions & 70 deletions data/evaluation/audio/interfaze-beta/eval_summary.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,8 +5,8 @@
"interfaze-beta"
],
"data_quality": {
"json_parse_fail_count": 0,
"json_non_structured_root_count": 0,
"json_parse_fail_count": 3,
"json_non_structured_root_count": 3,
"invalid_schema_input_count": 0,
"unknown_difficulty_count": 0,
"malformed_jsonl_line_count": 0
Expand All @@ -16,14 +16,14 @@
"n": 115,
"metrics": {
"json_parse_success": {
"mean": 1.0,
"ci95_low": 1.0,
"mean": 0.9739130434782609,
"ci95_low": 0.9391304347826087,
"ci95_high": 1.0,
"metric_name": "JSON Parse Success"
},
"json_root_structured": {
"mean": 1.0,
"ci95_low": 1.0,
"mean": 0.9739130434782609,
"ci95_low": 0.9391304347826087,
"ci95_high": 1.0,
"metric_name": "Structured JSON Root"
},
Expand All @@ -34,39 +34,39 @@
"metric_name": "Schema Valid Input"
},
"schema_compliance": {
"mean": 0.8434782608695652,
"ci95_low": 0.7739130434782608,
"ci95_high": 0.9043478260869565,
"mean": 0.8782608695652174,
"ci95_low": 0.8173913043478261,
"ci95_high": 0.9391304347826087,
"metric_name": "JSON Pass Rate"
},
"leaf_value_em": {
"mean": 0.2035126943660157,
"ci95_low": 0.17103765793432008,
"ci95_high": 0.23741275870060108,
"mean": 0.23077301234224104,
"ci95_low": 0.1930329140115948,
"ci95_high": 0.26680038850045346,
"metric_name": "Truth Score"
},
"value_token_f1": {
"mean": 0.40979085174131163,
"ci95_low": 0.3676184394443381,
"ci95_high": 0.45917501182305825,
"mean": 0.42478823082811235,
"ci95_low": 0.38225211308769336,
"ci95_high": 0.4745606015122529,
"metric_name": "Faithfulness Score"
},
"hier_path_recall": {
"mean": 0.7525844250377696,
"ci95_low": 0.6821627487702453,
"ci95_high": 0.8164828651631216,
"mean": 0.786259852042265,
"ci95_low": 0.7272657482603747,
"ci95_high": 0.8399056154183809,
"metric_name": "Path Recall"
},
"path_set_f1": {
"mean": 0.760510818069284,
"ci95_low": 0.6982341816918611,
"ci95_high": 0.8200800041652657,
"mean": 0.7937747480478013,
"ci95_low": 0.7331483471984981,
"ci95_high": 0.8500703888623989,
"metric_name": "Structure Coverage"
},
"type_precision": {
"mean": 0.8434782608695652,
"ci95_low": 0.7739130434782608,
"ci95_high": 0.9130434782608695,
"mean": 0.8782608695652174,
"ci95_low": 0.8173913043478261,
"ci95_high": 0.9304347826086956,
"metric_name": "Type Safety"
},
"strict_json_em": {
Expand All @@ -78,9 +78,9 @@
},
"category_scores": {
"Long Context Extraction": {
"mean": 0.455295990381699,
"ci95_low": 0.4151706968317004,
"ci95_high": 0.49527492797097805,
"mean": 0.48060703173753944,
"ci95_low": 0.4329209206336758,
"ci95_high": 0.5203834054329745,
"category_name": "Long Context Extraction",
"components": [
"leaf_value_em",
Expand All @@ -89,9 +89,9 @@
]
},
"Complex Schema Handling": {
"mean": 0.8158224466028048,
"ci95_low": 0.7481439044300071,
"ci95_high": 0.8759841683396631,
"mean": 0.850098829059412,
"ci95_low": 0.7860643146971029,
"ci95_high": 0.9022466825217262,
"category_name": "Complex Schema Handling",
"components": [
"schema_compliance",
Expand All @@ -100,19 +100,19 @@
]
},
"Multi-Context Linking": {
"mean": 0.30665177305366365,
"ci95_low": 0.2679415384215724,
"ci95_high": 0.344213751681674,
"mean": 0.32778062158517673,
"ci95_low": 0.28776454392138834,
"ci95_high": 0.3707548530060054,
"category_name": "Multi-Context Linking",
"components": [
"leaf_value_em",
"value_token_f1"
]
},
"Output Contract Reliability": {
"mean": 0.8956521739130435,
"ci95_low": 0.8434782608695652,
"ci95_high": 0.9362318840579711,
"mean": 0.9101449275362319,
"ci95_low": 0.8608695652173913,
"ci95_high": 0.9507246376811593,
"category_name": "Output Contract Reliability",
"components": [
"json_parse_success",
Expand All @@ -135,14 +135,14 @@
"n": 115,
"metrics": {
"json_parse_success": {
"mean": 1.0,
"ci95_low": 1.0,
"mean": 0.9737609329446064,
"ci95_low": 0.938953488372093,
"ci95_high": 1.0,
"metric_name": "JSON Parse Success"
},
"json_root_structured": {
"mean": 1.0,
"ci95_low": 1.0,
"mean": 0.9737609329446064,
"ci95_low": 0.938953488372093,
"ci95_high": 1.0,
"metric_name": "Structured JSON Root"
},
Expand All @@ -153,39 +153,39 @@
"metric_name": "Schema Valid Input"
},
"schema_compliance": {
"mean": 0.8425655976676385,
"ci95_low": 0.7732558139534884,
"ci95_high": 0.911504424778761,
"mean": 0.8775510204081632,
"ci95_low": 0.8168604651162791,
"ci95_high": 0.9302325581395349,
"metric_name": "JSON Pass Rate"
},
"leaf_value_em": {
"mean": 0.20469935730692537,
"ci95_low": 0.1704665631059015,
"ci95_high": 0.23875318937515516,
"mean": 0.22999829892361642,
"ci95_low": 0.19393893122745373,
"ci95_high": 0.2629245535249263,
"metric_name": "Truth Score"
},
"value_token_f1": {
"mean": 0.40870736953514497,
"ci95_low": 0.3627944278559024,
"ci95_high": 0.45182892624577414,
"mean": 0.4239177947321078,
"ci95_low": 0.37775592800979196,
"ci95_high": 0.47282847278461054,
"metric_name": "Faithfulness Score"
},
"hier_path_recall": {
"mean": 0.7511417686239956,
"ci95_low": 0.6884415972103335,
"ci95_high": 0.8087760153990141,
"mean": 0.7851669986400256,
"ci95_low": 0.7252075897884498,
"ci95_high": 0.8407465866635563,
"metric_name": "Path Recall"
},
"path_set_f1": {
"mean": 0.7594544642171425,
"ci95_low": 0.6913442023524712,
"ci95_high": 0.8199826794306149,
"mean": 0.7931067679586729,
"ci95_low": 0.736107466585722,
"ci95_high": 0.8417136971460728,
"metric_name": "Structure Coverage"
},
"type_precision": {
"mean": 0.8425655976676385,
"ci95_low": 0.7725947521865889,
"ci95_high": 0.9040697674418605,
"mean": 0.8775510204081632,
"ci95_low": 0.8081395348837209,
"ci95_high": 0.9384164222873901,
"metric_name": "Type Safety"
},
"strict_json_em": {
Expand All @@ -197,9 +197,9 @@
},
"category_scores": {
"Long Context Extraction": {
"mean": 0.45484949848868866,
"ci95_low": 0.4100811636782316,
"ci95_high": 0.5013095103464918,
"mean": 0.4796943640985833,
"ci95_low": 0.43891895392248975,
"ci95_high": 0.5205338105306229,
"category_name": "Long Context Extraction",
"components": [
"leaf_value_em",
Expand All @@ -208,9 +208,9 @@
]
},
"Complex Schema Handling": {
"mean": 0.8148618865174733,
"ci95_low": 0.7445595678806685,
"ci95_high": 0.872890555427766,
"mean": 0.8494029362583332,
"ci95_low": 0.7817889204122557,
"ci95_high": 0.9103566940095154,
"category_name": "Complex Schema Handling",
"components": [
"schema_compliance",
Expand All @@ -219,19 +219,19 @@
]
},
"Multi-Context Linking": {
"mean": 0.3067033634210352,
"ci95_low": 0.26983533749301075,
"ci95_high": 0.3447669368005712,
"mean": 0.32695804682786206,
"ci95_low": 0.2891042438311676,
"ci95_high": 0.36504533620428,
"category_name": "Multi-Context Linking",
"components": [
"leaf_value_em",
"value_token_f1"
]
},
"Output Contract Reliability": {
"mean": 0.8950437317784257,
"ci95_low": 0.847953216374269,
"ci95_high": 0.936046511627907,
"mean": 0.9096209912536443,
"ci95_low": 0.8604651162790697,
"ci95_high": 0.9505813953488372,
"category_name": "Output Contract Reliability",
"components": [
"json_parse_success",
Expand Down
Loading
Loading