{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:OTALFZJWSNXIX6P5J6UR2LNTXV","short_pith_number":"pith:OTALFZJW","canonical_record":{"source":{"id":"2502.20238","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-27T16:23:25Z","cross_cats_sorted":[],"title_canon_sha256":"1fe3f9fdc9e55266039dd6bbffb6421cac2b6a6df27318a1309db02f44fd7f1b","abstract_canon_sha256":"904a8de5f2ab7da35054d26e0b1d5e43dd8049121ba758995081505abf6099ae"},"schema_version":"1.0"},"canonical_sha256":"74c0b2e536936e8bf9fd4fa91d2db3bd72989be1f069240f2c3d0d08a1b82e7d","source":{"kind":"arxiv","id":"2502.20238","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.20238","created_at":"2026-07-05T11:13:21Z"},{"alias_kind":"arxiv_version","alias_value":"2502.20238v2","created_at":"2026-07-05T11:13:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20238","created_at":"2026-07-05T11:13:21Z"},{"alias_kind":"pith_short_12","alias_value":"OTALFZJWSNXI","created_at":"2026-07-05T11:13:21Z"},{"alias_kind":"pith_short_16","alias_value":"OTALFZJWSNXIX6P5","created_at":"2026-07-05T11:13:21Z"},{"alias_kind":"pith_short_8","alias_value":"OTALFZJW","created_at":"2026-07-05T11:13:21Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:OTALFZJWSNXIX6P5J6UR2LNTXV","target":"record","payload":{"canonical_record":{"source":{"id":"2502.20238","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-27T16:23:25Z","cross_cats_sorted":[],"title_canon_sha256":"1fe3f9fdc9e55266039dd6bbffb6421cac2b6a6df27318a1309db02f44fd7f1b","abstract_canon_sha256":"904a8de5f2ab7da35054d26e0b1d5e43dd8049121ba758995081505abf6099ae"},"schema_version":"1.0"},"canonical_sha256":"74c0b2e536936e8bf9fd4fa91d2db3bd72989be1f069240f2c3d0d08a1b82e7d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:13:21.587134Z","signature_b64":"TdtpTH/AKBV3Tx+CT//1RLJ3EzuhcVMDUw7FmM2mBCCMcEJbjZ1wGUCPdJGsichfaZvjvQ8dCsogUwQnwtPmCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"74c0b2e536936e8bf9fd4fa91d2db3bd72989be1f069240f2c3d0d08a1b82e7d","last_reissued_at":"2026-07-05T11:13:21.586651Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:13:21.586651Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.20238","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":"2026-07-05T11:13:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5XnGSErG6EECRE7a5208+czKfsNwcTGBq7rDAnyHxoV42Emrar12anGEhg/EW/3NZgGUqvqaLrKMuSNvHFkPDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T04:11:51.732288Z"},"content_sha256":"b1911a30c54d94d8828e0cb203f100436bfa6d4c8a9bda9f94eef6d8bbcec103","schema_version":"1.0","event_id":"sha256:b1911a30c54d94d8828e0cb203f100436bfa6d4c8a9bda9f94eef6d8bbcec103"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:OTALFZJWSNXIX6P5J6UR2LNTXV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"FINEREASON: Evaluating and Improving LLMs' Deliberate Reasoning through Reflective Puzzle Solving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Anh Tuan Luu, Chaoqun Liu, Deli Zhao, Guizhen Chen, Hao Zhang, Hou Pong Chan, Lidong Bing, Weiwen Xu, Yu Rong","submitted_at":"2025-02-27T16:23:25Z","abstract_excerpt":"Many challenging reasoning tasks require not just rapid, intuitive responses, but a more deliberate, multi-step approach. Recent progress in large language models (LLMs) highlights an important shift from the \"System 1\" way of quick reactions to the \"System 2\" style of reflection-and-correction problem solving. However, current benchmarks heavily rely on the final-answer accuracy, leaving much of a model's intermediate reasoning steps unexamined. This fails to assess the model's ability to reflect and rectify mistakes within the reasoning process. To bridge this gap, we introduce FINEREASON, a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20238","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/2502.20238/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":"2026-07-05T11:13:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"c5Kra4f4eMnrjfr9tyHq6utdsCalzQbCFknfYOWWIx8sujJxgfYm2uYPisa5ceASIeXeRyGOoTL/UjlGhT2HBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T04:11:51.732665Z"},"content_sha256":"74ab71f6ce25e785885e1a4436ef6cc66ae65d70741fad86bfb069f080f432f2","schema_version":"1.0","event_id":"sha256:74ab71f6ce25e785885e1a4436ef6cc66ae65d70741fad86bfb069f080f432f2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OTALFZJWSNXIX6P5J6UR2LNTXV/bundle.json","state_url":"https://pith.science/pith/OTALFZJWSNXIX6P5J6UR2LNTXV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OTALFZJWSNXIX6P5J6UR2LNTXV/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":"2026-07-07T04:11:51Z","links":{"resolver":"https://pith.science/pith/OTALFZJWSNXIX6P5J6UR2LNTXV","bundle":"https://pith.science/pith/OTALFZJWSNXIX6P5J6UR2LNTXV/bundle.json","state":"https://pith.science/pith/OTALFZJWSNXIX6P5J6UR2LNTXV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OTALFZJWSNXIX6P5J6UR2LNTXV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:OTALFZJWSNXIX6P5J6UR2LNTXV","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":"904a8de5f2ab7da35054d26e0b1d5e43dd8049121ba758995081505abf6099ae","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-27T16:23:25Z","title_canon_sha256":"1fe3f9fdc9e55266039dd6bbffb6421cac2b6a6df27318a1309db02f44fd7f1b"},"schema_version":"1.0","source":{"id":"2502.20238","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.20238","created_at":"2026-07-05T11:13:21Z"},{"alias_kind":"arxiv_version","alias_value":"2502.20238v2","created_at":"2026-07-05T11:13:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20238","created_at":"2026-07-05T11:13:21Z"},{"alias_kind":"pith_short_12","alias_value":"OTALFZJWSNXI","created_at":"2026-07-05T11:13:21Z"},{"alias_kind":"pith_short_16","alias_value":"OTALFZJWSNXIX6P5","created_at":"2026-07-05T11:13:21Z"},{"alias_kind":"pith_short_8","alias_value":"OTALFZJW","created_at":"2026-07-05T11:13:21Z"}],"graph_snapshots":[{"event_id":"sha256:74ab71f6ce25e785885e1a4436ef6cc66ae65d70741fad86bfb069f080f432f2","target":"graph","created_at":"2026-07-05T11:13:21Z","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/2502.20238/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Many challenging reasoning tasks require not just rapid, intuitive responses, but a more deliberate, multi-step approach. Recent progress in large language models (LLMs) highlights an important shift from the \"System 1\" way of quick reactions to the \"System 2\" style of reflection-and-correction problem solving. However, current benchmarks heavily rely on the final-answer accuracy, leaving much of a model's intermediate reasoning steps unexamined. This fails to assess the model's ability to reflect and rectify mistakes within the reasoning process. To bridge this gap, we introduce FINEREASON, a","authors_text":"Anh Tuan Luu, Chaoqun Liu, Deli Zhao, Guizhen Chen, Hao Zhang, Hou Pong Chan, Lidong Bing, Weiwen Xu, Yu Rong","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-27T16:23:25Z","title":"FINEREASON: Evaluating and Improving LLMs' Deliberate Reasoning through Reflective Puzzle Solving"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20238","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:b1911a30c54d94d8828e0cb203f100436bfa6d4c8a9bda9f94eef6d8bbcec103","target":"record","created_at":"2026-07-05T11:13:21Z","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":"904a8de5f2ab7da35054d26e0b1d5e43dd8049121ba758995081505abf6099ae","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-27T16:23:25Z","title_canon_sha256":"1fe3f9fdc9e55266039dd6bbffb6421cac2b6a6df27318a1309db02f44fd7f1b"},"schema_version":"1.0","source":{"id":"2502.20238","kind":"arxiv","version":2}},"canonical_sha256":"74c0b2e536936e8bf9fd4fa91d2db3bd72989be1f069240f2c3d0d08a1b82e7d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"74c0b2e536936e8bf9fd4fa91d2db3bd72989be1f069240f2c3d0d08a1b82e7d","first_computed_at":"2026-07-05T11:13:21.586651Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:13:21.586651Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"TdtpTH/AKBV3Tx+CT//1RLJ3EzuhcVMDUw7FmM2mBCCMcEJbjZ1wGUCPdJGsichfaZvjvQ8dCsogUwQnwtPmCg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:13:21.587134Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.20238","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b1911a30c54d94d8828e0cb203f100436bfa6d4c8a9bda9f94eef6d8bbcec103","sha256:74ab71f6ce25e785885e1a4436ef6cc66ae65d70741fad86bfb069f080f432f2"],"state_sha256":"f40e4bfb7de958b87ffb7adab2edc5c919a131ffa71136bb6e6dacb3a1c78225"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/MewJgzdhb4HMwekjfphT3KCjzGTK62J77MQEmcUu2L0VuUn/9f0Fm/BnW7HnFTv0XhVCvrqRxIsxuLON2CDAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T04:11:51.734615Z","bundle_sha256":"3b51945285d7f3e14358698a0baa694da811966080629a1df780cc101a11a143"}}