{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:MPRB2MFIHNB6V3YALTI4UTKRKT","short_pith_number":"pith:MPRB2MFI","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"},"canonical_sha256":"63e21d30a83b43eaef005cd1ca4d5154f3c2c522295ab8bf6ae2f5b05d00bd68","source":{"kind":"arxiv","id":"2601.04424","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2601.04424","created_at":"2026-07-02T00:18:26Z"},{"alias_kind":"arxiv_version","alias_value":"2601.04424v2","created_at":"2026-07-02T00:18:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.04424","created_at":"2026-07-02T00:18:26Z"},{"alias_kind":"pith_short_12","alias_value":"MPRB2MFIHNB6","created_at":"2026-07-02T00:18:26Z"},{"alias_kind":"pith_short_16","alias_value":"MPRB2MFIHNB6V3YA","created_at":"2026-07-02T00:18:26Z"},{"alias_kind":"pith_short_8","alias_value":"MPRB2MFI","created_at":"2026-07-02T00:18:26Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:MPRB2MFIHNB6V3YALTI4UTKRKT","target":"record","payload":{"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"},"canonical_sha256":"63e21d30a83b43eaef005cd1ca4d5154f3c2c522295ab8bf6ae2f5b05d00bd68","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"},"source_kind":"arxiv","source_id":"2601.04424","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-02T00:18:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vxNe3B2WNNPJCN5blhyM1c4gmVmOwgXgyCcIvtAmHVm5cOk9PN3qEj+gLysd882VCs7r4GhIMiMg3S6HeXPoAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T05:41:57.789947Z"},"content_sha256":"c2c34b0e3b97d75e02556c1455034f2c0711b6980c5800a7e88ac7c9a73068b0","schema_version":"1.0","event_id":"sha256:c2c34b0e3b97d75e02556c1455034f2c0711b6980c5800a7e88ac7c9a73068b0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:MPRB2MFIHNB6V3YALTI4UTKRKT","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-02T00:18:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0cYZehip4q0s664x2Fs7bBX7QdBqz3LplIIxfGmoI5NExmtwoxwgUq6Z2Dx6JVMo9HJXJ3VInI6evmDpkmfEAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T05:41:57.790572Z"},"content_sha256":"c51ea2b2cb1a6bee3e0e32e42cb949a6974cd6135caf9a674c39659c42d757b1","schema_version":"1.0","event_id":"sha256:c51ea2b2cb1a6bee3e0e32e42cb949a6974cd6135caf9a674c39659c42d757b1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/MPRB2MFIHNB6V3YALTI4UTKRKT/bundle.json","state_url":"https://pith.science/pith/MPRB2MFIHNB6V3YALTI4UTKRKT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/MPRB2MFIHNB6V3YALTI4UTKRKT/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-06T05:41:57Z","links":{"resolver":"https://pith.science/pith/MPRB2MFIHNB6V3YALTI4UTKRKT","bundle":"https://pith.science/pith/MPRB2MFIHNB6V3YALTI4UTKRKT/bundle.json","state":"https://pith.science/pith/MPRB2MFIHNB6V3YALTI4UTKRKT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/MPRB2MFIHNB6V3YALTI4UTKRKT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:MPRB2MFIHNB6V3YALTI4UTKRKT","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"80ffe6fb63f2c5eb65add1ccf0cfb28c3ae0f54478dc57d354a81c9bf2dfd994","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-01-07T22:08:17Z","title_canon_sha256":"eb10a5efb689c05a299e4baab837995cb68d93b3b7e715c8cb432f9e01923b7d"},"schema_version":"1.0","source":{"id":"2601.04424","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2601.04424","created_at":"2026-07-02T00:18:26Z"},{"alias_kind":"arxiv_version","alias_value":"2601.04424v2","created_at":"2026-07-02T00:18:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.04424","created_at":"2026-07-02T00:18:26Z"},{"alias_kind":"pith_short_12","alias_value":"MPRB2MFIHNB6","created_at":"2026-07-02T00:18:26Z"},{"alias_kind":"pith_short_16","alias_value":"MPRB2MFIHNB6V3YA","created_at":"2026-07-02T00:18:26Z"},{"alias_kind":"pith_short_8","alias_value":"MPRB2MFI","created_at":"2026-07-02T00:18:26Z"}],"graph_snapshots":[{"event_id":"sha256:c51ea2b2cb1a6bee3e0e32e42cb949a6974cd6135caf9a674c39659c42d757b1","target":"graph","created_at":"2026-07-02T00:18:26Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2601.04424/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Benjamin Mamut, Wei Xu, Yao Dou","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-01-07T22:08:17Z","title":"Gavel: Agent Meets Checklist for Evaluating LLMs on Long-Context Legal Summarization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.04424","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:c2c34b0e3b97d75e02556c1455034f2c0711b6980c5800a7e88ac7c9a73068b0","target":"record","created_at":"2026-07-02T00:18:26Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"80ffe6fb63f2c5eb65add1ccf0cfb28c3ae0f54478dc57d354a81c9bf2dfd994","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-01-07T22:08:17Z","title_canon_sha256":"eb10a5efb689c05a299e4baab837995cb68d93b3b7e715c8cb432f9e01923b7d"},"schema_version":"1.0","source":{"id":"2601.04424","kind":"arxiv","version":2}},"canonical_sha256":"63e21d30a83b43eaef005cd1ca4d5154f3c2c522295ab8bf6ae2f5b05d00bd68","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"63e21d30a83b43eaef005cd1ca4d5154f3c2c522295ab8bf6ae2f5b05d00bd68","first_computed_at":"2026-07-02T00:18:26.753121Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-02T00:18:26.753121Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"GQy9AglVs68DWDendHPKNXOdKyazuVxRFIeYKPwrWfqxQY0y51eLgOAU3SKSaoPwF75SZAYaJADJzTT7n6fRBQ==","signature_status":"signed_v1","signed_at":"2026-07-02T00:18:26.753934Z","signed_message":"canonical_sha256_bytes"},"source_id":"2601.04424","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c2c34b0e3b97d75e02556c1455034f2c0711b6980c5800a7e88ac7c9a73068b0","sha256:c51ea2b2cb1a6bee3e0e32e42cb949a6974cd6135caf9a674c39659c42d757b1"],"state_sha256":"b28cc9d9034c65d435669bedcc3a271239d09c4693d58a6f4959b1a3b5aaa695"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JX77io+UuVFSH9RfwikB1o6T+sI7CJq49Ps/bSApmOUXtPh/vrW+1F9thyDZZCOqD6FXrMq/lJiX8P/pZPjSBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T05:41:57.795622Z","bundle_sha256":"822fbde45893cfddf68be29f0b215401680df3d0ad19261ca5e7392cea9ec8c3"}}