{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DURIBKVQAEUHYITJLRRU2MTPC3","short_pith_number":"pith:DURIBKVQ","schema_version":"1.0","canonical_sha256":"1d2280aab001287c22695c634d326f16f92521ba868cb3fa88c7e4730ee8ee27","source":{"kind":"arxiv","id":"2606.22329","version":1},"attestation_state":"computed","paper":{"title":"BabelJudge: Measuring LLM-as-a-Judge Reliability Across Languages and Agent Trajectories","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Shreyas KC","submitted_at":"2026-06-21T04:35:43Z","abstract_excerpt":"LLM-as-a-judge has become the dominant approach to scalable evaluation in NLP pipelines, yet judges themselves carry systematic biases that raw accuracy hides: they favor responses placed in slot A (position bias), they prefer longer responses regardless of quality (verbosity bias), and their reliability degrades sharply in lower-resource languages. We introduce BabelJudge, an open-source benchmark and reliability audit framework that measures all four failure modes -- position bias, verbosity bias, order inconsistency, and cross-lingual degradation -- on any judge model, without requiring hum"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.22329","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-21T04:35:43Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"beef7822141189a0852b25af33ad7a6b9c81128789a9575c2a8e11ca05df6e7f","abstract_canon_sha256":"7c87cce20dcecd702fb59a578af577a2a0fcfe579bd38d825158322c382f1d4f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T02:13:34.996803Z","signature_b64":"/hHd6Y0+F4yNPL6z8bixEyKm48L4/dwJAG7v2OGamriWpYe1kxfUWw/UqwNE6hoKWeSJeXGxd0XcJlwtRQLADA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1d2280aab001287c22695c634d326f16f92521ba868cb3fa88c7e4730ee8ee27","last_reissued_at":"2026-06-23T02:13:34.996397Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T02:13:34.996397Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BabelJudge: Measuring LLM-as-a-Judge Reliability Across Languages and Agent Trajectories","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Shreyas KC","submitted_at":"2026-06-21T04:35:43Z","abstract_excerpt":"LLM-as-a-judge has become the dominant approach to scalable evaluation in NLP pipelines, yet judges themselves carry systematic biases that raw accuracy hides: they favor responses placed in slot A (position bias), they prefer longer responses regardless of quality (verbosity bias), and their reliability degrades sharply in lower-resource languages. We introduce BabelJudge, an open-source benchmark and reliability audit framework that measures all four failure modes -- position bias, verbosity bias, order inconsistency, and cross-lingual degradation -- on any judge model, without requiring hum"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.22329","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2606.22329/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.22329","created_at":"2026-06-23T02:13:34.996459+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.22329v1","created_at":"2026-06-23T02:13:34.996459+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.22329","created_at":"2026-06-23T02:13:34.996459+00:00"},{"alias_kind":"pith_short_12","alias_value":"DURIBKVQAEUH","created_at":"2026-06-23T02:13:34.996459+00:00"},{"alias_kind":"pith_short_16","alias_value":"DURIBKVQAEUHYITJ","created_at":"2026-06-23T02:13:34.996459+00:00"},{"alias_kind":"pith_short_8","alias_value":"DURIBKVQ","created_at":"2026-06-23T02:13:34.996459+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DURIBKVQAEUHYITJLRRU2MTPC3","json":"https://pith.science/pith/DURIBKVQAEUHYITJLRRU2MTPC3.json","graph_json":"https://pith.science/api/pith-number/DURIBKVQAEUHYITJLRRU2MTPC3/graph.json","events_json":"https://pith.science/api/pith-number/DURIBKVQAEUHYITJLRRU2MTPC3/events.json","paper":"https://pith.science/paper/DURIBKVQ"},"agent_actions":{"view_html":"https://pith.science/pith/DURIBKVQAEUHYITJLRRU2MTPC3","download_json":"https://pith.science/pith/DURIBKVQAEUHYITJLRRU2MTPC3.json","view_paper":"https://pith.science/paper/DURIBKVQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.22329&json=true","fetch_graph":"https://pith.science/api/pith-number/DURIBKVQAEUHYITJLRRU2MTPC3/graph.json","fetch_events":"https://pith.science/api/pith-number/DURIBKVQAEUHYITJLRRU2MTPC3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DURIBKVQAEUHYITJLRRU2MTPC3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DURIBKVQAEUHYITJLRRU2MTPC3/action/storage_attestation","attest_author":"https://pith.science/pith/DURIBKVQAEUHYITJLRRU2MTPC3/action/author_attestation","sign_citation":"https://pith.science/pith/DURIBKVQAEUHYITJLRRU2MTPC3/action/citation_signature","submit_replication":"https://pith.science/pith/DURIBKVQAEUHYITJLRRU2MTPC3/action/replication_record"}},"created_at":"2026-06-23T02:13:34.996459+00:00","updated_at":"2026-06-23T02:13:34.996459+00:00"}