{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:MA5GLWWYKZVARUTGX4HI6CYFFI","short_pith_number":"pith:MA5GLWWY","canonical_record":{"source":{"id":"2508.15361","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-21T08:43:35Z","cross_cats_sorted":[],"title_canon_sha256":"9e93627e07a757f1982b253835f51e472b59c515fb3a76e72838fde2fa11f273","abstract_canon_sha256":"3d80125a5b06a98bdc19c363ae99313271b353b4485a5aa9c3ecf4d7eafff9b2"},"schema_version":"1.0"},"canonical_sha256":"603a65dad8566a08d266bf0e8f0b052a2c0177cd346721c5fa5c7876b9521af8","source":{"kind":"arxiv","id":"2508.15361","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.15361","created_at":"2026-07-05T11:57:10Z"},{"alias_kind":"arxiv_version","alias_value":"2508.15361v1","created_at":"2026-07-05T11:57:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.15361","created_at":"2026-07-05T11:57:10Z"},{"alias_kind":"pith_short_12","alias_value":"MA5GLWWYKZVA","created_at":"2026-07-05T11:57:10Z"},{"alias_kind":"pith_short_16","alias_value":"MA5GLWWYKZVARUTG","created_at":"2026-07-05T11:57:10Z"},{"alias_kind":"pith_short_8","alias_value":"MA5GLWWY","created_at":"2026-07-05T11:57:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:MA5GLWWYKZVARUTGX4HI6CYFFI","target":"record","payload":{"canonical_record":{"source":{"id":"2508.15361","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-21T08:43:35Z","cross_cats_sorted":[],"title_canon_sha256":"9e93627e07a757f1982b253835f51e472b59c515fb3a76e72838fde2fa11f273","abstract_canon_sha256":"3d80125a5b06a98bdc19c363ae99313271b353b4485a5aa9c3ecf4d7eafff9b2"},"schema_version":"1.0"},"canonical_sha256":"603a65dad8566a08d266bf0e8f0b052a2c0177cd346721c5fa5c7876b9521af8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:57:10.913388Z","signature_b64":"w6VaxZvC34ImcWgF/VjHZ7qzWZ9DYYhOnXjKrm90+YWvcGt0RlCb+NS3oDtinBolQaO90DwbvdXajpN/bTI9Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"603a65dad8566a08d266bf0e8f0b052a2c0177cd346721c5fa5c7876b9521af8","last_reissued_at":"2026-07-05T11:57:10.912908Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:57:10.912908Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.15361","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:57:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PfVr1EWtcSarDdyR/sAKrNSQiaUkRRoI4QcgRZQP050Iege92eyZRI1ej0BBHAug2O9ec51nBpn39lWig4j/Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T17:19:15.179585Z"},"content_sha256":"4621c33a3758a05d48e1ee24472cd94758464cc7b2fcfe48a059c9fe0e3256cd","schema_version":"1.0","event_id":"sha256:4621c33a3758a05d48e1ee24472cd94758464cc7b2fcfe48a059c9fe0e3256cd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:MA5GLWWYKZVARUTGX4HI6CYFFI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Survey on Large Language Model Benchmarks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bingli Wang, Chengming Li, Guhong Chen, Le Sun, Liyang Fan, Min Yang, Qiyao Wang, Ruifeng Xu, Shiwen Ni, Shuaimin Li, Siyi Li, Xingjian Wang, Xuanang Chen, Yifan Zhang","submitted_at":"2025-08-21T08:43:35Z","abstract_excerpt":"In recent years, with the rapid development of the depth and breadth of large language models' capabilities, various corresponding evaluation benchmarks have been emerging in increasing numbers. As a quantitative assessment tool for model performance, benchmarks are not only a core means to measure model capabilities but also a key element in guiding the direction of model development and promoting technological innovation. We systematically review the current status and development of large language model benchmarks for the first time, categorizing 283 representative benchmarks into three cat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.15361","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/2508.15361/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":"2026-07-05T11:57:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WbXgIlhRk7Vl0tXdMZ9lVdYBztphbMnQ5yiXN9VsSU95C6b3DNHXrz3Fb1+g01XtQDjvp1h2MczSo7O5T+u7Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T17:19:15.180653Z"},"content_sha256":"602b4a8fd49fd9c8a275b978c21aa0d24647b9ad4f29e3d050db9dd6d8c687d0","schema_version":"1.0","event_id":"sha256:602b4a8fd49fd9c8a275b978c21aa0d24647b9ad4f29e3d050db9dd6d8c687d0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/MA5GLWWYKZVARUTGX4HI6CYFFI/bundle.json","state_url":"https://pith.science/pith/MA5GLWWYKZVARUTGX4HI6CYFFI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/MA5GLWWYKZVARUTGX4HI6CYFFI/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":"2026-08-05T17:19:15Z","links":{"resolver":"https://pith.science/pith/MA5GLWWYKZVARUTGX4HI6CYFFI","bundle":"https://pith.science/pith/MA5GLWWYKZVARUTGX4HI6CYFFI/bundle.json","state":"https://pith.science/pith/MA5GLWWYKZVARUTGX4HI6CYFFI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/MA5GLWWYKZVARUTGX4HI6CYFFI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:MA5GLWWYKZVARUTGX4HI6CYFFI","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":"3d80125a5b06a98bdc19c363ae99313271b353b4485a5aa9c3ecf4d7eafff9b2","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-21T08:43:35Z","title_canon_sha256":"9e93627e07a757f1982b253835f51e472b59c515fb3a76e72838fde2fa11f273"},"schema_version":"1.0","source":{"id":"2508.15361","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.15361","created_at":"2026-07-05T11:57:10Z"},{"alias_kind":"arxiv_version","alias_value":"2508.15361v1","created_at":"2026-07-05T11:57:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.15361","created_at":"2026-07-05T11:57:10Z"},{"alias_kind":"pith_short_12","alias_value":"MA5GLWWYKZVA","created_at":"2026-07-05T11:57:10Z"},{"alias_kind":"pith_short_16","alias_value":"MA5GLWWYKZVARUTG","created_at":"2026-07-05T11:57:10Z"},{"alias_kind":"pith_short_8","alias_value":"MA5GLWWY","created_at":"2026-07-05T11:57:10Z"}],"graph_snapshots":[{"event_id":"sha256:602b4a8fd49fd9c8a275b978c21aa0d24647b9ad4f29e3d050db9dd6d8c687d0","target":"graph","created_at":"2026-07-05T11:57:10Z","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/2508.15361/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In recent years, with the rapid development of the depth and breadth of large language models' capabilities, various corresponding evaluation benchmarks have been emerging in increasing numbers. As a quantitative assessment tool for model performance, benchmarks are not only a core means to measure model capabilities but also a key element in guiding the direction of model development and promoting technological innovation. We systematically review the current status and development of large language model benchmarks for the first time, categorizing 283 representative benchmarks into three cat","authors_text":"Bingli Wang, Chengming Li, Guhong Chen, Le Sun, Liyang Fan, Min Yang, Qiyao Wang, Ruifeng Xu, Shiwen Ni, Shuaimin Li, Siyi Li, Xingjian Wang, Xuanang Chen, Yifan Zhang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-21T08:43:35Z","title":"A Survey on Large Language Model Benchmarks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.15361","kind":"arxiv","version":1},"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":"sha256:4621c33a3758a05d48e1ee24472cd94758464cc7b2fcfe48a059c9fe0e3256cd","target":"record","created_at":"2026-07-05T11:57:10Z","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":"3d80125a5b06a98bdc19c363ae99313271b353b4485a5aa9c3ecf4d7eafff9b2","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-21T08:43:35Z","title_canon_sha256":"9e93627e07a757f1982b253835f51e472b59c515fb3a76e72838fde2fa11f273"},"schema_version":"1.0","source":{"id":"2508.15361","kind":"arxiv","version":1}},"canonical_sha256":"603a65dad8566a08d266bf0e8f0b052a2c0177cd346721c5fa5c7876b9521af8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"603a65dad8566a08d266bf0e8f0b052a2c0177cd346721c5fa5c7876b9521af8","first_computed_at":"2026-07-05T11:57:10.912908Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:57:10.912908Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"w6VaxZvC34ImcWgF/VjHZ7qzWZ9DYYhOnXjKrm90+YWvcGt0RlCb+NS3oDtinBolQaO90DwbvdXajpN/bTI9Dg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:57:10.913388Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.15361","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4621c33a3758a05d48e1ee24472cd94758464cc7b2fcfe48a059c9fe0e3256cd","sha256:602b4a8fd49fd9c8a275b978c21aa0d24647b9ad4f29e3d050db9dd6d8c687d0"],"state_sha256":"d846eb477e334288943053211dc0d087b2fe7aa0b8c26af5d06058a0f3a3eb72"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VIOj9TIJWQf4U3gQnILuwfEY7cORVvB2dqeIeJsZFy4x+f8mTc5xJRBOOgUc2YloK8Qz4McPBkfmOWSlAj6JBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T17:19:15.187931Z","bundle_sha256":"df4d159f0966b547da9333f1cafe579d145815c66a73a26882cac68f440d1880"}}