{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HIF7TB4RKOZIOEA6C2KBWC2FLE","short_pith_number":"pith:HIF7TB4R","schema_version":"1.0","canonical_sha256":"3a0bf9879153b287101e16941b0b45590e1b68f7b8643b84f4ef721dcccf66fc","source":{"kind":"arxiv","id":"2402.11756","version":3},"attestation_state":"computed","paper":{"title":"MARS: Meaning-Aware Response Scoring for Uncertainty Estimation in Generative LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Baturalp Buyukates, Chenyang Tao, Dimitrios Dimitriadis, Duygu Nur Yaldiz, Salman Avestimehr, Yavuz Faruk Bakman","submitted_at":"2024-02-19T01:04:22Z","abstract_excerpt":"Generative Large Language Models (LLMs) are widely utilized for their excellence in various tasks. However, their tendency to produce inaccurate or misleading outputs poses a potential risk, particularly in high-stakes environments. Therefore, estimating the correctness of generative LLM outputs is an important task for enhanced reliability. Uncertainty Estimation (UE) in generative LLMs is an evolving domain, where SOTA probability-based methods commonly employ length-normalized scoring. In this work, we propose Meaning-Aware Response Scoring (MARS) as an alternative to length-normalized scor"},"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":"2402.11756","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T01:04:22Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"9fc4f02a1c53a0b09188224ddfc1aa567913d68e99c7501afa52040e64e660b4","abstract_canon_sha256":"7a22918afca3a84f5d66006ec4a1616acfbc65e87cca1b22b90c2617b8e7e59c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:13:26.251529Z","signature_b64":"5PME128LAg0mib7Ch5xxSkOGtMiqFl74nc/MCj+k4uuI7BzNke4/VrB6glgFvvU+4QD2jQRHKlF5ZAc8pfm9Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a0bf9879153b287101e16941b0b45590e1b68f7b8643b84f4ef721dcccf66fc","last_reissued_at":"2026-07-05T10:13:26.251038Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:13:26.251038Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MARS: Meaning-Aware Response Scoring for Uncertainty Estimation in Generative LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Baturalp Buyukates, Chenyang Tao, Dimitrios Dimitriadis, Duygu Nur Yaldiz, Salman Avestimehr, Yavuz Faruk Bakman","submitted_at":"2024-02-19T01:04:22Z","abstract_excerpt":"Generative Large Language Models (LLMs) are widely utilized for their excellence in various tasks. However, their tendency to produce inaccurate or misleading outputs poses a potential risk, particularly in high-stakes environments. Therefore, estimating the correctness of generative LLM outputs is an important task for enhanced reliability. Uncertainty Estimation (UE) in generative LLMs is an evolving domain, where SOTA probability-based methods commonly employ length-normalized scoring. In this work, we propose Meaning-Aware Response Scoring (MARS) as an alternative to length-normalized scor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.11756","kind":"arxiv","version":3},"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/2402.11756/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":"2402.11756","created_at":"2026-07-05T10:13:26.251095+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.11756v3","created_at":"2026-07-05T10:13:26.251095+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.11756","created_at":"2026-07-05T10:13:26.251095+00:00"},{"alias_kind":"pith_short_12","alias_value":"HIF7TB4RKOZI","created_at":"2026-07-05T10:13:26.251095+00:00"},{"alias_kind":"pith_short_16","alias_value":"HIF7TB4RKOZIOEA6","created_at":"2026-07-05T10:13:26.251095+00:00"},{"alias_kind":"pith_short_8","alias_value":"HIF7TB4R","created_at":"2026-07-05T10:13:26.251095+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.17200","citing_title":"Calibrating Model-Based Evaluation Metrics for Summarization","ref_index":96,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HIF7TB4RKOZIOEA6C2KBWC2FLE","json":"https://pith.science/pith/HIF7TB4RKOZIOEA6C2KBWC2FLE.json","graph_json":"https://pith.science/api/pith-number/HIF7TB4RKOZIOEA6C2KBWC2FLE/graph.json","events_json":"https://pith.science/api/pith-number/HIF7TB4RKOZIOEA6C2KBWC2FLE/events.json","paper":"https://pith.science/paper/HIF7TB4R"},"agent_actions":{"view_html":"https://pith.science/pith/HIF7TB4RKOZIOEA6C2KBWC2FLE","download_json":"https://pith.science/pith/HIF7TB4RKOZIOEA6C2KBWC2FLE.json","view_paper":"https://pith.science/paper/HIF7TB4R","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.11756&json=true","fetch_graph":"https://pith.science/api/pith-number/HIF7TB4RKOZIOEA6C2KBWC2FLE/graph.json","fetch_events":"https://pith.science/api/pith-number/HIF7TB4RKOZIOEA6C2KBWC2FLE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HIF7TB4RKOZIOEA6C2KBWC2FLE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HIF7TB4RKOZIOEA6C2KBWC2FLE/action/storage_attestation","attest_author":"https://pith.science/pith/HIF7TB4RKOZIOEA6C2KBWC2FLE/action/author_attestation","sign_citation":"https://pith.science/pith/HIF7TB4RKOZIOEA6C2KBWC2FLE/action/citation_signature","submit_replication":"https://pith.science/pith/HIF7TB4RKOZIOEA6C2KBWC2FLE/action/replication_record"}},"created_at":"2026-07-05T10:13:26.251095+00:00","updated_at":"2026-07-05T10:13:26.251095+00:00"}