{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:NLMF3BJFWWE53UUNEKEV3VR4UK","short_pith_number":"pith:NLMF3BJF","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"},"canonical_sha256":"6ad85d8525b589ddd28d22895dd63ca295e44e5af66637fc9b804069665337d0","source":{"kind":"arxiv","id":"2505.19293","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.19293","created_at":"2026-06-04T01:08:29Z"},{"alias_kind":"arxiv_version","alias_value":"2505.19293v2","created_at":"2026-06-04T01:08:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.19293","created_at":"2026-06-04T01:08:29Z"},{"alias_kind":"pith_short_12","alias_value":"NLMF3BJFWWE5","created_at":"2026-06-04T01:08:29Z"},{"alias_kind":"pith_short_16","alias_value":"NLMF3BJFWWE53UUN","created_at":"2026-06-04T01:08:29Z"},{"alias_kind":"pith_short_8","alias_value":"NLMF3BJF","created_at":"2026-06-04T01:08:29Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:NLMF3BJFWWE53UUNEKEV3VR4UK","target":"record","payload":{"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"},"canonical_sha256":"6ad85d8525b589ddd28d22895dd63ca295e44e5af66637fc9b804069665337d0","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"},"source_kind":"arxiv","source_id":"2505.19293","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":"2026-06-04T01:08:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UG+AUxFOqUERu0vHzUevKTd/hsw/LesxMNuLvkRoiEXRVxYnG8ut4Sxu/YdR2WXQSSL5hmUEs5nylXRM87tiCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T12:36:36.009136Z"},"content_sha256":"6599b24313e2a00ac0dca07037948a954fcf09c8ae126c095967642281ceefaf","schema_version":"1.0","event_id":"sha256:6599b24313e2a00ac0dca07037948a954fcf09c8ae126c095967642281ceefaf"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:NLMF3BJFWWE53UUNEKEV3VR4UK","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-04T01:08:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mpppx6+RsDMea8u0B7R+OxIXQH9I0o+434sORP0PevlaXk/MnskpKDaWr3utLF18tMFNnM18VVEMLPVNWJFEAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T12:36:36.010028Z"},"content_sha256":"441da74617bdec7b3dfc38bf571160b078177d0b4b576fb3fa5b8f0784d7aa39","schema_version":"1.0","event_id":"sha256:441da74617bdec7b3dfc38bf571160b078177d0b4b576fb3fa5b8f0784d7aa39"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NLMF3BJFWWE53UUNEKEV3VR4UK/bundle.json","state_url":"https://pith.science/pith/NLMF3BJFWWE53UUNEKEV3VR4UK/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NLMF3BJFWWE53UUNEKEV3VR4UK/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":"2026-06-10T12:36:36Z","links":{"resolver":"https://pith.science/pith/NLMF3BJFWWE53UUNEKEV3VR4UK","bundle":"https://pith.science/pith/NLMF3BJFWWE53UUNEKEV3VR4UK/bundle.json","state":"https://pith.science/pith/NLMF3BJFWWE53UUNEKEV3VR4UK/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NLMF3BJFWWE53UUNEKEV3VR4UK/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:NLMF3BJFWWE53UUNEKEV3VR4UK","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":"c3c96cdc0f7d98114a9930f7d8faf526cf382b51190e98184ae466f362b4e59a","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-25T19:58:31Z","title_canon_sha256":"a99aeb24122fcda0a21bc9f9148f36630055bbb885458018b06b819cd71d585a"},"schema_version":"1.0","source":{"id":"2505.19293","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.19293","created_at":"2026-06-04T01:08:29Z"},{"alias_kind":"arxiv_version","alias_value":"2505.19293v2","created_at":"2026-06-04T01:08:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.19293","created_at":"2026-06-04T01:08:29Z"},{"alias_kind":"pith_short_12","alias_value":"NLMF3BJFWWE5","created_at":"2026-06-04T01:08:29Z"},{"alias_kind":"pith_short_16","alias_value":"NLMF3BJFWWE53UUN","created_at":"2026-06-04T01:08:29Z"},{"alias_kind":"pith_short_8","alias_value":"NLMF3BJF","created_at":"2026-06-04T01:08:29Z"}],"graph_snapshots":[{"event_id":"sha256:441da74617bdec7b3dfc38bf571160b078177d0b4b576fb3fa5b8f0784d7aa39","target":"graph","created_at":"2026-06-04T01:08:29Z","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/2505.19293/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Hongye Jin, Qifan Wang, Shaochen Zhong, Song Jiang, Vipin Chaudhary, Wang Yang, Xiaotian Han","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-25T19:58:31Z","title":"100-LongBench: Are de facto Long-Context Benchmarks Literally Evaluating Long-Context Ability?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.19293","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":"sha256:6599b24313e2a00ac0dca07037948a954fcf09c8ae126c095967642281ceefaf","target":"record","created_at":"2026-06-04T01:08:29Z","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":"c3c96cdc0f7d98114a9930f7d8faf526cf382b51190e98184ae466f362b4e59a","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-25T19:58:31Z","title_canon_sha256":"a99aeb24122fcda0a21bc9f9148f36630055bbb885458018b06b819cd71d585a"},"schema_version":"1.0","source":{"id":"2505.19293","kind":"arxiv","version":2}},"canonical_sha256":"6ad85d8525b589ddd28d22895dd63ca295e44e5af66637fc9b804069665337d0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6ad85d8525b589ddd28d22895dd63ca295e44e5af66637fc9b804069665337d0","first_computed_at":"2026-06-04T01:08:29.323040Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-04T01:08:29.323040Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mdTrjEkhWdYsZcwQW47vxhiyZ8Y1g7zS/xP2E6pafu6B1gQhtOHffpXTCpsVTTJioTn9ORnSBTzYZbFH4UdHAA==","signature_status":"signed_v1","signed_at":"2026-06-04T01:08:29.323589Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.19293","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6599b24313e2a00ac0dca07037948a954fcf09c8ae126c095967642281ceefaf","sha256:441da74617bdec7b3dfc38bf571160b078177d0b4b576fb3fa5b8f0784d7aa39"],"state_sha256":"798ad03405a5f4955ad389dd4180c34799d302fccd63d131a9a99066ae6ccca9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RKJcMHlN/eOFNnVz8xoN+dC4y2EDdwa1FCcYmpf9ZbookNvaKjN2egKTT55YJ3n/MOqFlDvRYRg+kiq91ZhsCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-10T12:36:36.015083Z","bundle_sha256":"a5e45e6e86c44d4f11601792a522b203dd2b8b9098d974f62322c5f0bcc47ade"}}