{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:XRNI2SIUNN62PUNTY22QYMZ2BJ","short_pith_number":"pith:XRNI2SIU","canonical_record":{"source":{"id":"2508.19363","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-26T18:41:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2299180b97c8def0b3c103a638d9d89e750ac0900969f834fb7161fe40d25e3b","abstract_canon_sha256":"7261eade552f2546ef10f5ec1c1f2efbfe8fe89485795b558acef66665bf3607"},"schema_version":"1.0"},"canonical_sha256":"bc5a8d49146b7da7d1b3c6b50c333a0a6944e6188e389fc1fd62b0d43daf3304","source":{"kind":"arxiv","id":"2508.19363","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.19363","created_at":"2026-07-05T11:59:44Z"},{"alias_kind":"arxiv_version","alias_value":"2508.19363v1","created_at":"2026-07-05T11:59:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.19363","created_at":"2026-07-05T11:59:44Z"},{"alias_kind":"pith_short_12","alias_value":"XRNI2SIUNN62","created_at":"2026-07-05T11:59:44Z"},{"alias_kind":"pith_short_16","alias_value":"XRNI2SIUNN62PUNT","created_at":"2026-07-05T11:59:44Z"},{"alias_kind":"pith_short_8","alias_value":"XRNI2SIU","created_at":"2026-07-05T11:59:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:XRNI2SIUNN62PUNTY22QYMZ2BJ","target":"record","payload":{"canonical_record":{"source":{"id":"2508.19363","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-26T18:41:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2299180b97c8def0b3c103a638d9d89e750ac0900969f834fb7161fe40d25e3b","abstract_canon_sha256":"7261eade552f2546ef10f5ec1c1f2efbfe8fe89485795b558acef66665bf3607"},"schema_version":"1.0"},"canonical_sha256":"bc5a8d49146b7da7d1b3c6b50c333a0a6944e6188e389fc1fd62b0d43daf3304","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:59:44.131301Z","signature_b64":"HzG5Ykaxz3xRHnTRSuipTGl/4f/7BnpHjzcGIiDdHUgf2hOa7qpBTIMlvA/sVoG1/5sIsoS2eU9YDWGmGKrNDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bc5a8d49146b7da7d1b3c6b50c333a0a6944e6188e389fc1fd62b0d43daf3304","last_reissued_at":"2026-07-05T11:59:44.130718Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:59:44.130718Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.19363","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":"2026-07-05T11:59:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Rl7nV61LlmaVokTRzk4J5JrmSdMG9sC38m5u2F3rl1Hu8FuAS/XXQES/qqGZ9avXJAkhH/5fQvs+a8Y+IoFeBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T13:53:51.095253Z"},"content_sha256":"244ad77651082fe0867f3a8230209e6e459aed7736d0683f8814961a9ab48741","schema_version":"1.0","event_id":"sha256:244ad77651082fe0867f3a8230209e6e459aed7736d0683f8814961a9ab48741"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:XRNI2SIUNN62PUNTY22QYMZ2BJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LongReasonArena: A Long Reasoning Benchmark for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Furu Wei, Jiayu Ding, Lei Cui, Nanning Zheng, Shuming Ma","submitted_at":"2025-08-26T18:41:53Z","abstract_excerpt":"Existing long-context benchmarks for Large Language Models (LLMs) focus on evaluating comprehension of long inputs, while overlooking the evaluation of long reasoning abilities. To address this gap, we introduce LongReasonArena, a benchmark specifically designed to assess the long reasoning capabilities of LLMs. Our tasks require models to solve problems by executing multi-step algorithms that reflect key aspects of long reasoning, such as retrieval and backtracking. By controlling the inputs, the required reasoning length can be arbitrarily scaled, reaching up to 1 million tokens of reasoning"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.19363","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/2508.19363/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":"2026-07-05T11:59:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QsFr4+U3/xHJC2DWa0nJQ4CST9C+HzNW2+7ESyDj5vK/CZ0Blms1u/88zienaY7vIniB2OMWp41hKBFkFksqBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T13:53:51.096204Z"},"content_sha256":"6c23baf721295a12402dc4a7f89f53ae3c17cccfe42c92cc566e619f62e947a9","schema_version":"1.0","event_id":"sha256:6c23baf721295a12402dc4a7f89f53ae3c17cccfe42c92cc566e619f62e947a9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XRNI2SIUNN62PUNTY22QYMZ2BJ/bundle.json","state_url":"https://pith.science/pith/XRNI2SIUNN62PUNTY22QYMZ2BJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XRNI2SIUNN62PUNTY22QYMZ2BJ/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":"2026-08-06T13:53:51Z","links":{"resolver":"https://pith.science/pith/XRNI2SIUNN62PUNTY22QYMZ2BJ","bundle":"https://pith.science/pith/XRNI2SIUNN62PUNTY22QYMZ2BJ/bundle.json","state":"https://pith.science/pith/XRNI2SIUNN62PUNTY22QYMZ2BJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XRNI2SIUNN62PUNTY22QYMZ2BJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:XRNI2SIUNN62PUNTY22QYMZ2BJ","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":"7261eade552f2546ef10f5ec1c1f2efbfe8fe89485795b558acef66665bf3607","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-26T18:41:53Z","title_canon_sha256":"2299180b97c8def0b3c103a638d9d89e750ac0900969f834fb7161fe40d25e3b"},"schema_version":"1.0","source":{"id":"2508.19363","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.19363","created_at":"2026-07-05T11:59:44Z"},{"alias_kind":"arxiv_version","alias_value":"2508.19363v1","created_at":"2026-07-05T11:59:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.19363","created_at":"2026-07-05T11:59:44Z"},{"alias_kind":"pith_short_12","alias_value":"XRNI2SIUNN62","created_at":"2026-07-05T11:59:44Z"},{"alias_kind":"pith_short_16","alias_value":"XRNI2SIUNN62PUNT","created_at":"2026-07-05T11:59:44Z"},{"alias_kind":"pith_short_8","alias_value":"XRNI2SIU","created_at":"2026-07-05T11:59:44Z"}],"graph_snapshots":[{"event_id":"sha256:6c23baf721295a12402dc4a7f89f53ae3c17cccfe42c92cc566e619f62e947a9","target":"graph","created_at":"2026-07-05T11:59:44Z","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/2508.19363/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Existing long-context benchmarks for Large Language Models (LLMs) focus on evaluating comprehension of long inputs, while overlooking the evaluation of long reasoning abilities. To address this gap, we introduce LongReasonArena, a benchmark specifically designed to assess the long reasoning capabilities of LLMs. Our tasks require models to solve problems by executing multi-step algorithms that reflect key aspects of long reasoning, such as retrieval and backtracking. By controlling the inputs, the required reasoning length can be arbitrarily scaled, reaching up to 1 million tokens of reasoning","authors_text":"Furu Wei, Jiayu Ding, Lei Cui, Nanning Zheng, Shuming Ma","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-26T18:41:53Z","title":"LongReasonArena: A Long Reasoning Benchmark for Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.19363","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":"sha256:244ad77651082fe0867f3a8230209e6e459aed7736d0683f8814961a9ab48741","target":"record","created_at":"2026-07-05T11:59:44Z","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":"7261eade552f2546ef10f5ec1c1f2efbfe8fe89485795b558acef66665bf3607","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-26T18:41:53Z","title_canon_sha256":"2299180b97c8def0b3c103a638d9d89e750ac0900969f834fb7161fe40d25e3b"},"schema_version":"1.0","source":{"id":"2508.19363","kind":"arxiv","version":1}},"canonical_sha256":"bc5a8d49146b7da7d1b3c6b50c333a0a6944e6188e389fc1fd62b0d43daf3304","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bc5a8d49146b7da7d1b3c6b50c333a0a6944e6188e389fc1fd62b0d43daf3304","first_computed_at":"2026-07-05T11:59:44.130718Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:59:44.130718Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HzG5Ykaxz3xRHnTRSuipTGl/4f/7BnpHjzcGIiDdHUgf2hOa7qpBTIMlvA/sVoG1/5sIsoS2eU9YDWGmGKrNDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:59:44.131301Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.19363","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:244ad77651082fe0867f3a8230209e6e459aed7736d0683f8814961a9ab48741","sha256:6c23baf721295a12402dc4a7f89f53ae3c17cccfe42c92cc566e619f62e947a9"],"state_sha256":"6dee08ea051fffa2d96650e70bd48d36a591ce1903c1d2487978252144f423dd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lBNvoxauO5XF4relWYq8uBg476QmQAVHfOyanl2EhH/FF0vFzjN4IlcEME0cLspnecN6yHSAZd5DTf0AnGf5Aw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T13:53:51.102965Z","bundle_sha256":"ffeae57461a3b24c854dee729c64f7574b213480240a8ef6e19ce6bbcab25f80"}}