{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:NLMF3BJFWWE53UUNEKEV3VR4UK","short_pith_number":"pith:NLMF3BJF","schema_version":"1.0","canonical_sha256":"6ad85d8525b589ddd28d22895dd63ca295e44e5af66637fc9b804069665337d0","source":{"kind":"arxiv","id":"2505.19293","version":2},"attestation_state":"computed","paper":{"title":"100-LongBench: Are de facto Long-Context Benchmarks Literally Evaluating Long-Context Ability?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Hongye Jin, Qifan Wang, Shaochen Zhong, Song Jiang, Vipin Chaudhary, Wang Yang, Xiaotian Han","submitted_at":"2025-05-25T19:58:31Z","abstract_excerpt":"Long-context capability is considered one of the most important abilities of LLMs, as a truly long context-capable LLM enables users to effortlessly process many originally exhausting tasks -- e.g., digesting a long-form document to find answers vs. directly asking an LLM about it. However, existing real-task-based long-context evaluation benchmarks have two major shortcomings. First, benchmarks like LongBench often do not provide proper metrics to separate long-context performance from the model's baseline ability, making cross-model comparison unclear. Second, such benchmarks are usually con"},"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":"2505.19293","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-25T19:58:31Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"a99aeb24122fcda0a21bc9f9148f36630055bbb885458018b06b819cd71d585a","abstract_canon_sha256":"c3c96cdc0f7d98114a9930f7d8faf526cf382b51190e98184ae466f362b4e59a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-04T01:08:29.323589Z","signature_b64":"mdTrjEkhWdYsZcwQW47vxhiyZ8Y1g7zS/xP2E6pafu6B1gQhtOHffpXTCpsVTTJioTn9ORnSBTzYZbFH4UdHAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6ad85d8525b589ddd28d22895dd63ca295e44e5af66637fc9b804069665337d0","last_reissued_at":"2026-06-04T01:08:29.323040Z","signature_status":"signed_v1","first_computed_at":"2026-06-04T01:08:29.323040Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"100-LongBench: Are de facto Long-Context Benchmarks Literally Evaluating Long-Context Ability?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Hongye Jin, Qifan Wang, Shaochen Zhong, Song Jiang, Vipin Chaudhary, Wang Yang, Xiaotian Han","submitted_at":"2025-05-25T19:58:31Z","abstract_excerpt":"Long-context capability is considered one of the most important abilities of LLMs, as a truly long context-capable LLM enables users to effortlessly process many originally exhausting tasks -- e.g., digesting a long-form document to find answers vs. directly asking an LLM about it. However, existing real-task-based long-context evaluation benchmarks have two major shortcomings. First, benchmarks like LongBench often do not provide proper metrics to separate long-context performance from the model's baseline ability, making cross-model comparison unclear. Second, such benchmarks are usually con"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.19293","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/2505.19293/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":"2505.19293","created_at":"2026-06-04T01:08:29.323108+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.19293v2","created_at":"2026-06-04T01:08:29.323108+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.19293","created_at":"2026-06-04T01:08:29.323108+00:00"},{"alias_kind":"pith_short_12","alias_value":"NLMF3BJFWWE5","created_at":"2026-06-04T01:08:29.323108+00:00"},{"alias_kind":"pith_short_16","alias_value":"NLMF3BJFWWE53UUN","created_at":"2026-06-04T01:08:29.323108+00:00"},{"alias_kind":"pith_short_8","alias_value":"NLMF3BJF","created_at":"2026-06-04T01:08:29.323108+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.23170","citing_title":"Positional Failures in Long-Context LLMs: A Blind Spot in Reasoning Benchmarks","ref_index":32,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NLMF3BJFWWE53UUNEKEV3VR4UK","json":"https://pith.science/pith/NLMF3BJFWWE53UUNEKEV3VR4UK.json","graph_json":"https://pith.science/api/pith-number/NLMF3BJFWWE53UUNEKEV3VR4UK/graph.json","events_json":"https://pith.science/api/pith-number/NLMF3BJFWWE53UUNEKEV3VR4UK/events.json","paper":"https://pith.science/paper/NLMF3BJF"},"agent_actions":{"view_html":"https://pith.science/pith/NLMF3BJFWWE53UUNEKEV3VR4UK","download_json":"https://pith.science/pith/NLMF3BJFWWE53UUNEKEV3VR4UK.json","view_paper":"https://pith.science/paper/NLMF3BJF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.19293&json=true","fetch_graph":"https://pith.science/api/pith-number/NLMF3BJFWWE53UUNEKEV3VR4UK/graph.json","fetch_events":"https://pith.science/api/pith-number/NLMF3BJFWWE53UUNEKEV3VR4UK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NLMF3BJFWWE53UUNEKEV3VR4UK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NLMF3BJFWWE53UUNEKEV3VR4UK/action/storage_attestation","attest_author":"https://pith.science/pith/NLMF3BJFWWE53UUNEKEV3VR4UK/action/author_attestation","sign_citation":"https://pith.science/pith/NLMF3BJFWWE53UUNEKEV3VR4UK/action/citation_signature","submit_replication":"https://pith.science/pith/NLMF3BJFWWE53UUNEKEV3VR4UK/action/replication_record"}},"created_at":"2026-06-04T01:08:29.323108+00:00","updated_at":"2026-06-04T01:08:29.323108+00:00"}