{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:2JFGQVOQ57FIZ26FVI7QSUD5VU","short_pith_number":"pith:2JFGQVOQ","canonical_record":{"source":{"id":"2404.10346","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-16T07:30:11Z","cross_cats_sorted":[],"title_canon_sha256":"006cce95b1a70372739183657286943d5d67d2e5ce373d9f02a234960f35879b","abstract_canon_sha256":"0b7f1adf1294302053617fcfd54ef67c6054b121f18185b226f6f01eb0e8b101"},"schema_version":"1.0"},"canonical_sha256":"d24a6855d0efca8cebc5aa3f09507dad0132faad6d1cebc16d97cdcf432d565d","source":{"kind":"arxiv","id":"2404.10346","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.10346","created_at":"2026-07-05T09:15:09Z"},{"alias_kind":"arxiv_version","alias_value":"2404.10346v4","created_at":"2026-07-05T09:15:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.10346","created_at":"2026-07-05T09:15:09Z"},{"alias_kind":"pith_short_12","alias_value":"2JFGQVOQ57FI","created_at":"2026-07-05T09:15:09Z"},{"alias_kind":"pith_short_16","alias_value":"2JFGQVOQ57FIZ26F","created_at":"2026-07-05T09:15:09Z"},{"alias_kind":"pith_short_8","alias_value":"2JFGQVOQ","created_at":"2026-07-05T09:15:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:2JFGQVOQ57FIZ26FVI7QSUD5VU","target":"record","payload":{"canonical_record":{"source":{"id":"2404.10346","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-16T07:30:11Z","cross_cats_sorted":[],"title_canon_sha256":"006cce95b1a70372739183657286943d5d67d2e5ce373d9f02a234960f35879b","abstract_canon_sha256":"0b7f1adf1294302053617fcfd54ef67c6054b121f18185b226f6f01eb0e8b101"},"schema_version":"1.0"},"canonical_sha256":"d24a6855d0efca8cebc5aa3f09507dad0132faad6d1cebc16d97cdcf432d565d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:15:09.850955Z","signature_b64":"qdzeHMv0InF334irj7JnyHmBR5bWhiirUorOl4d9XoJmjjA43OPnhXMvlUH0A0WvJ26ZhU8yrvbYDgn6j9P2Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d24a6855d0efca8cebc5aa3f09507dad0132faad6d1cebc16d97cdcf432d565d","last_reissued_at":"2026-07-05T09:15:09.850491Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:15:09.850491Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2404.10346","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:15:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CApskwGZDBvhGl1K186m2vgzZ3eGZZN6JHgNFCbBkShYjOOvpiCV3O8WTuDnTstxFrdcTiUSt+oqYe4EOkpQDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T05:12:35.177269Z"},"content_sha256":"432a3493acb07ed6e0d96e6e78a24498d42c5ff31106b50fa06105d0faa700ff","schema_version":"1.0","event_id":"sha256:432a3493acb07ed6e0d96e6e78a24498d42c5ff31106b50fa06105d0faa700ff"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:2JFGQVOQ57FIZ26FVI7QSUD5VU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Self-Explore: Enhancing Mathematical Reasoning in Language Models with Fine-grained Rewards","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Doyoung Kim, Hyeonbin Hwang, Minjoon Seo, Seonghyeon Ye, Seungone Kim","submitted_at":"2024-04-16T07:30:11Z","abstract_excerpt":"Training on large amounts of rationales (i.e., CoT Fine-tuning) is effective at improving the reasoning capabilities of large language models (LLMs). However, acquiring human-authored rationales or augmenting rationales from proprietary models is costly and not scalable. In this paper, we study the problem of whether LLMs could self-improve their reasoning capabilities. To this end, we propose Self-Explore, where the LLM is tasked to explore the first wrong step (i.e., the first pit) within the rationale and use such signals as fine-grained rewards for further improvement. On the GSM8K and MAT"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.10346","kind":"arxiv","version":4},"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/2404.10346/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":"2026-07-05T09:15:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WFOI2AizGX7mPNWiDacXuileUpswTgftKsgAovvgr734YAGOpCSexeohKLp7axS3WjiWRLiD/EIWz1jgFoJFBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T05:12:35.177654Z"},"content_sha256":"43ab9c3c70db0849669d6c436a096f2732a27a3c2e01caad744a3263b9dddab8","schema_version":"1.0","event_id":"sha256:43ab9c3c70db0849669d6c436a096f2732a27a3c2e01caad744a3263b9dddab8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/2JFGQVOQ57FIZ26FVI7QSUD5VU/bundle.json","state_url":"https://pith.science/pith/2JFGQVOQ57FIZ26FVI7QSUD5VU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/2JFGQVOQ57FIZ26FVI7QSUD5VU/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":"2026-08-07T05:12:35Z","links":{"resolver":"https://pith.science/pith/2JFGQVOQ57FIZ26FVI7QSUD5VU","bundle":"https://pith.science/pith/2JFGQVOQ57FIZ26FVI7QSUD5VU/bundle.json","state":"https://pith.science/pith/2JFGQVOQ57FIZ26FVI7QSUD5VU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/2JFGQVOQ57FIZ26FVI7QSUD5VU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:2JFGQVOQ57FIZ26FVI7QSUD5VU","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":"0b7f1adf1294302053617fcfd54ef67c6054b121f18185b226f6f01eb0e8b101","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-16T07:30:11Z","title_canon_sha256":"006cce95b1a70372739183657286943d5d67d2e5ce373d9f02a234960f35879b"},"schema_version":"1.0","source":{"id":"2404.10346","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.10346","created_at":"2026-07-05T09:15:09Z"},{"alias_kind":"arxiv_version","alias_value":"2404.10346v4","created_at":"2026-07-05T09:15:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.10346","created_at":"2026-07-05T09:15:09Z"},{"alias_kind":"pith_short_12","alias_value":"2JFGQVOQ57FI","created_at":"2026-07-05T09:15:09Z"},{"alias_kind":"pith_short_16","alias_value":"2JFGQVOQ57FIZ26F","created_at":"2026-07-05T09:15:09Z"},{"alias_kind":"pith_short_8","alias_value":"2JFGQVOQ","created_at":"2026-07-05T09:15:09Z"}],"graph_snapshots":[{"event_id":"sha256:43ab9c3c70db0849669d6c436a096f2732a27a3c2e01caad744a3263b9dddab8","target":"graph","created_at":"2026-07-05T09:15:09Z","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/2404.10346/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Training on large amounts of rationales (i.e., CoT Fine-tuning) is effective at improving the reasoning capabilities of large language models (LLMs). However, acquiring human-authored rationales or augmenting rationales from proprietary models is costly and not scalable. In this paper, we study the problem of whether LLMs could self-improve their reasoning capabilities. To this end, we propose Self-Explore, where the LLM is tasked to explore the first wrong step (i.e., the first pit) within the rationale and use such signals as fine-grained rewards for further improvement. On the GSM8K and MAT","authors_text":"Doyoung Kim, Hyeonbin Hwang, Minjoon Seo, Seonghyeon Ye, Seungone Kim","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-16T07:30:11Z","title":"Self-Explore: Enhancing Mathematical Reasoning in Language Models with Fine-grained Rewards"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.10346","kind":"arxiv","version":4},"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":"sha256:432a3493acb07ed6e0d96e6e78a24498d42c5ff31106b50fa06105d0faa700ff","target":"record","created_at":"2026-07-05T09:15:09Z","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":"0b7f1adf1294302053617fcfd54ef67c6054b121f18185b226f6f01eb0e8b101","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-16T07:30:11Z","title_canon_sha256":"006cce95b1a70372739183657286943d5d67d2e5ce373d9f02a234960f35879b"},"schema_version":"1.0","source":{"id":"2404.10346","kind":"arxiv","version":4}},"canonical_sha256":"d24a6855d0efca8cebc5aa3f09507dad0132faad6d1cebc16d97cdcf432d565d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d24a6855d0efca8cebc5aa3f09507dad0132faad6d1cebc16d97cdcf432d565d","first_computed_at":"2026-07-05T09:15:09.850491Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:15:09.850491Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"qdzeHMv0InF334irj7JnyHmBR5bWhiirUorOl4d9XoJmjjA43OPnhXMvlUH0A0WvJ26ZhU8yrvbYDgn6j9P2Cg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:15:09.850955Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.10346","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:432a3493acb07ed6e0d96e6e78a24498d42c5ff31106b50fa06105d0faa700ff","sha256:43ab9c3c70db0849669d6c436a096f2732a27a3c2e01caad744a3263b9dddab8"],"state_sha256":"ad1f1b7a8d248e0ff1a95ebafbe22330815ccda41df3c9748de2949a3a302eca"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6OKVe4SIzZc8mZeDumz/ENNYSozCQ31Tu8H6cCs6Tpzx7Ir7WPCqJaq4CyILEubIomfKPNCWLZ5gk96CnBOcCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T05:12:35.180580Z","bundle_sha256":"f01f0d694f7cc2c75e7acce4fa9fbf279e6e9adf95deb5ff811c0aee23a91c38"}}