{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:6GMXH2LGBJ7BX3MSIGBZH6O7HR","short_pith_number":"pith:6GMXH2LG","canonical_record":{"source":{"id":"2305.17306","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-26T23:46:42Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"6b621f1087f1635f439cd897b70dd9e52ac1acbc2863b9524ad0e54a1a04750e","abstract_canon_sha256":"b8a3b1ea9dbe7d4caa275e3c4dd7f341683bda1a59aa5634696d0f60ea5dd111"},"schema_version":"1.0"},"canonical_sha256":"f19973e9660a7e1bed92418393f9df3c646dccf38cda4f4e21c5347ae1f3228d","source":{"kind":"arxiv","id":"2305.17306","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.17306","created_at":"2026-07-05T06:14:26Z"},{"alias_kind":"arxiv_version","alias_value":"2305.17306v1","created_at":"2026-07-05T06:14:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.17306","created_at":"2026-07-05T06:14:26Z"},{"alias_kind":"pith_short_12","alias_value":"6GMXH2LGBJ7B","created_at":"2026-07-05T06:14:26Z"},{"alias_kind":"pith_short_16","alias_value":"6GMXH2LGBJ7BX3MS","created_at":"2026-07-05T06:14:26Z"},{"alias_kind":"pith_short_8","alias_value":"6GMXH2LG","created_at":"2026-07-05T06:14:26Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:6GMXH2LGBJ7BX3MSIGBZH6O7HR","target":"record","payload":{"canonical_record":{"source":{"id":"2305.17306","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-26T23:46:42Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"6b621f1087f1635f439cd897b70dd9e52ac1acbc2863b9524ad0e54a1a04750e","abstract_canon_sha256":"b8a3b1ea9dbe7d4caa275e3c4dd7f341683bda1a59aa5634696d0f60ea5dd111"},"schema_version":"1.0"},"canonical_sha256":"f19973e9660a7e1bed92418393f9df3c646dccf38cda4f4e21c5347ae1f3228d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:14:26.250418Z","signature_b64":"6tRZT0rs+Pgmd4w97UDWIFEDSfJCobG2SlpxNM9an+NZKYGU0mmbXCFBzKUsZqXjs4SDpNkOHZ6LJ6IDGM0VAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f19973e9660a7e1bed92418393f9df3c646dccf38cda4f4e21c5347ae1f3228d","last_reissued_at":"2026-07-05T06:14:26.249943Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:14:26.249943Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2305.17306","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":"2026-07-05T06:14:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"l95EuyzmEv2qkLb4QdLkDVq796g3jiNJJcSlORYw739YTA7D+M2Rb1YJT5yc1l3AhoNQ7t5rS2P0s/UyjtadBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-22T18:58:44.290008Z"},"content_sha256":"b3b66ee20ba8dabfc96c7daacaaf574f107b8e68e56f808fdd64bcea65ee99d9","schema_version":"1.0","event_id":"sha256:b3b66ee20ba8dabfc96c7daacaaf574f107b8e68e56f808fdd64bcea65ee99d9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:6GMXH2LGBJ7BX3MSIGBZH6O7HR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Chain-of-Thought Hub: A Continuous Effort to Measure Large Language Models' Reasoning Performance","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Hao Peng, Litu Ou, Mingyu Chen, Tushar Khot, Yao Fu, Yuhao Wan","submitted_at":"2023-05-26T23:46:42Z","abstract_excerpt":"As large language models (LLMs) are continuously being developed, their evaluation becomes increasingly important yet challenging. This work proposes Chain-of-Thought Hub, an open-source evaluation suite on the multi-step reasoning capabilities of large language models. We are interested in this setting for two reasons: (1) from the behavior of GPT and PaLM model family, we observe that complex reasoning is likely to be a key differentiator between weaker and stronger LLMs; (2) we envisage large language models to become the next-generation computational platform and foster an ecosystem of LLM"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.17306","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/2305.17306/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":"2026-07-05T06:14:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gfSwPkBmE/UoCsA8t+m+piRfDBcm+auMRikyLR31Qztm3HFW4GgBQW/Tec54KoElpTCtWjTe6lou7Di9JBziCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-22T18:58:44.290528Z"},"content_sha256":"8655ad4fb5e320ee77fd78f144b2b568a13081f27c5de4d6bae9929e630d1f23","schema_version":"1.0","event_id":"sha256:8655ad4fb5e320ee77fd78f144b2b568a13081f27c5de4d6bae9929e630d1f23"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6GMXH2LGBJ7BX3MSIGBZH6O7HR/bundle.json","state_url":"https://pith.science/pith/6GMXH2LGBJ7BX3MSIGBZH6O7HR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6GMXH2LGBJ7BX3MSIGBZH6O7HR/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":"2026-08-22T18:58:44Z","links":{"resolver":"https://pith.science/pith/6GMXH2LGBJ7BX3MSIGBZH6O7HR","bundle":"https://pith.science/pith/6GMXH2LGBJ7BX3MSIGBZH6O7HR/bundle.json","state":"https://pith.science/pith/6GMXH2LGBJ7BX3MSIGBZH6O7HR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6GMXH2LGBJ7BX3MSIGBZH6O7HR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:6GMXH2LGBJ7BX3MSIGBZH6O7HR","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":"b8a3b1ea9dbe7d4caa275e3c4dd7f341683bda1a59aa5634696d0f60ea5dd111","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-26T23:46:42Z","title_canon_sha256":"6b621f1087f1635f439cd897b70dd9e52ac1acbc2863b9524ad0e54a1a04750e"},"schema_version":"1.0","source":{"id":"2305.17306","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.17306","created_at":"2026-07-05T06:14:26Z"},{"alias_kind":"arxiv_version","alias_value":"2305.17306v1","created_at":"2026-07-05T06:14:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.17306","created_at":"2026-07-05T06:14:26Z"},{"alias_kind":"pith_short_12","alias_value":"6GMXH2LGBJ7B","created_at":"2026-07-05T06:14:26Z"},{"alias_kind":"pith_short_16","alias_value":"6GMXH2LGBJ7BX3MS","created_at":"2026-07-05T06:14:26Z"},{"alias_kind":"pith_short_8","alias_value":"6GMXH2LG","created_at":"2026-07-05T06:14:26Z"}],"graph_snapshots":[{"event_id":"sha256:8655ad4fb5e320ee77fd78f144b2b568a13081f27c5de4d6bae9929e630d1f23","target":"graph","created_at":"2026-07-05T06:14:26Z","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/2305.17306/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As large language models (LLMs) are continuously being developed, their evaluation becomes increasingly important yet challenging. This work proposes Chain-of-Thought Hub, an open-source evaluation suite on the multi-step reasoning capabilities of large language models. We are interested in this setting for two reasons: (1) from the behavior of GPT and PaLM model family, we observe that complex reasoning is likely to be a key differentiator between weaker and stronger LLMs; (2) we envisage large language models to become the next-generation computational platform and foster an ecosystem of LLM","authors_text":"Hao Peng, Litu Ou, Mingyu Chen, Tushar Khot, Yao Fu, Yuhao Wan","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-26T23:46:42Z","title":"Chain-of-Thought Hub: A Continuous Effort to Measure Large Language Models' Reasoning Performance"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.17306","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:b3b66ee20ba8dabfc96c7daacaaf574f107b8e68e56f808fdd64bcea65ee99d9","target":"record","created_at":"2026-07-05T06:14:26Z","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":"b8a3b1ea9dbe7d4caa275e3c4dd7f341683bda1a59aa5634696d0f60ea5dd111","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-26T23:46:42Z","title_canon_sha256":"6b621f1087f1635f439cd897b70dd9e52ac1acbc2863b9524ad0e54a1a04750e"},"schema_version":"1.0","source":{"id":"2305.17306","kind":"arxiv","version":1}},"canonical_sha256":"f19973e9660a7e1bed92418393f9df3c646dccf38cda4f4e21c5347ae1f3228d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f19973e9660a7e1bed92418393f9df3c646dccf38cda4f4e21c5347ae1f3228d","first_computed_at":"2026-07-05T06:14:26.249943Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:14:26.249943Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6tRZT0rs+Pgmd4w97UDWIFEDSfJCobG2SlpxNM9an+NZKYGU0mmbXCFBzKUsZqXjs4SDpNkOHZ6LJ6IDGM0VAA==","signature_status":"signed_v1","signed_at":"2026-07-05T06:14:26.250418Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.17306","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b3b66ee20ba8dabfc96c7daacaaf574f107b8e68e56f808fdd64bcea65ee99d9","sha256:8655ad4fb5e320ee77fd78f144b2b568a13081f27c5de4d6bae9929e630d1f23"],"state_sha256":"9c7edb94e4b5e67acfcc249ce162f6cd798c79a68834cee3eca6fb46432763ac"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"m/NRjsIK580yYpCZLsxEHPIHehydazS8LWapjAHc71knElG4iFGgNOnd7d+1tf+ekCs3yanpO+GAQkKii8PyAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-22T18:58:44.294569Z","bundle_sha256":"3ad3acdb0e764caff694ed6caed63b02ffe7eef3c794f0347848ae52071f302c"}}