{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:MPRB2MFIHNB6V3YALTI4UTKRKT","short_pith_number":"pith:MPRB2MFI","schema_version":"1.0","canonical_sha256":"63e21d30a83b43eaef005cd1ca4d5154f3c2c522295ab8bf6ae2f5b05d00bd68","source":{"kind":"arxiv","id":"2601.04424","version":2},"attestation_state":"computed","paper":{"title":"Gavel: Agent Meets Checklist for Evaluating LLMs on Long-Context Legal Summarization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Benjamin Mamut, Wei Xu, Yao Dou","submitted_at":"2026-01-07T22:08:17Z","abstract_excerpt":"Large language models (LLMs) now support contexts of up to 1M tokens, but their strengths and weaknesses on complex long-context tasks remain unclear. To study this, we focus on multi-document legal case summarization, where a single case often spans many documents exceeding 100K tokens. We systematically evaluate 12 frontier LLMs with Gavel, which consists of Gavel-Ref, a reference-based evaluation framework with checklist, residual-fact, and writing-style evaluations, and Gavel-Agent, a reference-free agent for evaluating factual coverage directly from source documents. Our results show that"},"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":"2601.04424","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-01-07T22:08:17Z","cross_cats_sorted":[],"title_canon_sha256":"eb10a5efb689c05a299e4baab837995cb68d93b3b7e715c8cb432f9e01923b7d","abstract_canon_sha256":"80ffe6fb63f2c5eb65add1ccf0cfb28c3ae0f54478dc57d354a81c9bf2dfd994"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-02T00:18:26.753934Z","signature_b64":"GQy9AglVs68DWDendHPKNXOdKyazuVxRFIeYKPwrWfqxQY0y51eLgOAU3SKSaoPwF75SZAYaJADJzTT7n6fRBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"63e21d30a83b43eaef005cd1ca4d5154f3c2c522295ab8bf6ae2f5b05d00bd68","last_reissued_at":"2026-07-02T00:18:26.753121Z","signature_status":"signed_v1","first_computed_at":"2026-07-02T00:18:26.753121Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Gavel: Agent Meets Checklist for Evaluating LLMs on Long-Context Legal Summarization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Benjamin Mamut, Wei Xu, Yao Dou","submitted_at":"2026-01-07T22:08:17Z","abstract_excerpt":"Large language models (LLMs) now support contexts of up to 1M tokens, but their strengths and weaknesses on complex long-context tasks remain unclear. To study this, we focus on multi-document legal case summarization, where a single case often spans many documents exceeding 100K tokens. We systematically evaluate 12 frontier LLMs with Gavel, which consists of Gavel-Ref, a reference-based evaluation framework with checklist, residual-fact, and writing-style evaluations, and Gavel-Agent, a reference-free agent for evaluating factual coverage directly from source documents. Our results show that"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.04424","kind":"arxiv","version":2},"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/2601.04424/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":"2601.04424","created_at":"2026-07-02T00:18:26.753234+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.04424v2","created_at":"2026-07-02T00:18:26.753234+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.04424","created_at":"2026-07-02T00:18:26.753234+00:00"},{"alias_kind":"pith_short_12","alias_value":"MPRB2MFIHNB6","created_at":"2026-07-02T00:18:26.753234+00:00"},{"alias_kind":"pith_short_16","alias_value":"MPRB2MFIHNB6V3YA","created_at":"2026-07-02T00:18:26.753234+00:00"},{"alias_kind":"pith_short_8","alias_value":"MPRB2MFI","created_at":"2026-07-02T00:18:26.753234+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/MPRB2MFIHNB6V3YALTI4UTKRKT","json":"https://pith.science/pith/MPRB2MFIHNB6V3YALTI4UTKRKT.json","graph_json":"https://pith.science/api/pith-number/MPRB2MFIHNB6V3YALTI4UTKRKT/graph.json","events_json":"https://pith.science/api/pith-number/MPRB2MFIHNB6V3YALTI4UTKRKT/events.json","paper":"https://pith.science/paper/MPRB2MFI"},"agent_actions":{"view_html":"https://pith.science/pith/MPRB2MFIHNB6V3YALTI4UTKRKT","download_json":"https://pith.science/pith/MPRB2MFIHNB6V3YALTI4UTKRKT.json","view_paper":"https://pith.science/paper/MPRB2MFI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.04424&json=true","fetch_graph":"https://pith.science/api/pith-number/MPRB2MFIHNB6V3YALTI4UTKRKT/graph.json","fetch_events":"https://pith.science/api/pith-number/MPRB2MFIHNB6V3YALTI4UTKRKT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MPRB2MFIHNB6V3YALTI4UTKRKT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MPRB2MFIHNB6V3YALTI4UTKRKT/action/storage_attestation","attest_author":"https://pith.science/pith/MPRB2MFIHNB6V3YALTI4UTKRKT/action/author_attestation","sign_citation":"https://pith.science/pith/MPRB2MFIHNB6V3YALTI4UTKRKT/action/citation_signature","submit_replication":"https://pith.science/pith/MPRB2MFIHNB6V3YALTI4UTKRKT/action/replication_record"}},"created_at":"2026-07-02T00:18:26.753234+00:00","updated_at":"2026-07-02T00:18:26.753234+00:00"}