{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:6LTZIJQXQ67KVDJJCN6AS4RGSK","short_pith_number":"pith:6LTZIJQX","canonical_record":{"source":{"id":"2505.24273","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-30T06:49:00Z","cross_cats_sorted":[],"title_canon_sha256":"4e48f92558ab3814e21a43e6972645e95f1d9e708a2bac053372d078a6d6a3c2","abstract_canon_sha256":"660e90bf69942a87a964461f11cdf32cfbb3cc1275083f9e90acbc2b51593b29"},"schema_version":"1.0"},"canonical_sha256":"f2e794261787beaa8d29137c097226928d2f2bdd55d7f38a957aa7a8004a4d47","source":{"kind":"arxiv","id":"2505.24273","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.24273","created_at":"2026-07-05T11:12:45Z"},{"alias_kind":"arxiv_version","alias_value":"2505.24273v1","created_at":"2026-07-05T11:12:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24273","created_at":"2026-07-05T11:12:45Z"},{"alias_kind":"pith_short_12","alias_value":"6LTZIJQXQ67K","created_at":"2026-07-05T11:12:45Z"},{"alias_kind":"pith_short_16","alias_value":"6LTZIJQXQ67KVDJJ","created_at":"2026-07-05T11:12:45Z"},{"alias_kind":"pith_short_8","alias_value":"6LTZIJQX","created_at":"2026-07-05T11:12:45Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:6LTZIJQXQ67KVDJJCN6AS4RGSK","target":"record","payload":{"canonical_record":{"source":{"id":"2505.24273","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-30T06:49:00Z","cross_cats_sorted":[],"title_canon_sha256":"4e48f92558ab3814e21a43e6972645e95f1d9e708a2bac053372d078a6d6a3c2","abstract_canon_sha256":"660e90bf69942a87a964461f11cdf32cfbb3cc1275083f9e90acbc2b51593b29"},"schema_version":"1.0"},"canonical_sha256":"f2e794261787beaa8d29137c097226928d2f2bdd55d7f38a957aa7a8004a4d47","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:45.696911Z","signature_b64":"vYbH4g6GgaDMmJQWcj0iVW1IHrdunGZcbkJQKc+ze9kGVwuysfsXPhWm5Ha0fhHb0sc3j/XSdwoJHF7zgynrCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f2e794261787beaa8d29137c097226928d2f2bdd55d7f38a957aa7a8004a4d47","last_reissued_at":"2026-07-05T11:12:45.696313Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:45.696313Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.24273","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":"2026-07-05T11:12:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8RsXU1CEH7nZYcUWIRlGQnYKhz23CyaiG7PvcAAwrjsvWBJbXDuULZ2QaqnfbAZYwxKh3/yZJTIWcVrPrdXLDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T09:25:53.454879Z"},"content_sha256":"4d187615e9ec9888b062cdc4b3e854abc01b486e75895e9bf609e4f0796ea852","schema_version":"1.0","event_id":"sha256:4d187615e9ec9888b062cdc4b3e854abc01b486e75895e9bf609e4f0796ea852"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:6LTZIJQXQ67KVDJJCN6AS4RGSK","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"How Much Backtracking is Enough? Exploring the Interplay of SFT and RL in Enhancing LLM Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bhuwan Dhingra, Hongyi James Cai, Junlin Wang, Xiaoyin Chen","submitted_at":"2025-05-30T06:49:00Z","abstract_excerpt":"Recent breakthroughs in large language models (LLMs) have effectively improved their reasoning abilities, particularly on mathematical and logical problems that have verifiable answers, through techniques such as supervised finetuning (SFT) and reinforcement learning (RL). Prior research indicates that RL effectively internalizes search strategies, enabling long chain-of-thought (CoT) reasoning, with backtracking emerging naturally as a learned capability. However, the precise benefits of backtracking, specifically, how significantly it contributes to reasoning improvements and the optimal ext"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24273","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/2505.24273/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":"2026-07-05T11:12:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ucEtyOtkM1o4TJxT0e2wTFXGb4E9o/BYRLgtqa223kLM7kAT+x3uksZR7RZKlWqe1EGOEKA/9yJ0IJc2KA6NCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T09:25:53.455439Z"},"content_sha256":"865d926069c85e6b7b75471756d6d69a11bf9c498927ac90aea12f2f3e81fe9f","schema_version":"1.0","event_id":"sha256:865d926069c85e6b7b75471756d6d69a11bf9c498927ac90aea12f2f3e81fe9f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6LTZIJQXQ67KVDJJCN6AS4RGSK/bundle.json","state_url":"https://pith.science/pith/6LTZIJQXQ67KVDJJCN6AS4RGSK/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6LTZIJQXQ67KVDJJCN6AS4RGSK/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":"2026-08-04T09:25:53Z","links":{"resolver":"https://pith.science/pith/6LTZIJQXQ67KVDJJCN6AS4RGSK","bundle":"https://pith.science/pith/6LTZIJQXQ67KVDJJCN6AS4RGSK/bundle.json","state":"https://pith.science/pith/6LTZIJQXQ67KVDJJCN6AS4RGSK/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6LTZIJQXQ67KVDJJCN6AS4RGSK/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:6LTZIJQXQ67KVDJJCN6AS4RGSK","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":"660e90bf69942a87a964461f11cdf32cfbb3cc1275083f9e90acbc2b51593b29","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-30T06:49:00Z","title_canon_sha256":"4e48f92558ab3814e21a43e6972645e95f1d9e708a2bac053372d078a6d6a3c2"},"schema_version":"1.0","source":{"id":"2505.24273","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.24273","created_at":"2026-07-05T11:12:45Z"},{"alias_kind":"arxiv_version","alias_value":"2505.24273v1","created_at":"2026-07-05T11:12:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24273","created_at":"2026-07-05T11:12:45Z"},{"alias_kind":"pith_short_12","alias_value":"6LTZIJQXQ67K","created_at":"2026-07-05T11:12:45Z"},{"alias_kind":"pith_short_16","alias_value":"6LTZIJQXQ67KVDJJ","created_at":"2026-07-05T11:12:45Z"},{"alias_kind":"pith_short_8","alias_value":"6LTZIJQX","created_at":"2026-07-05T11:12:45Z"}],"graph_snapshots":[{"event_id":"sha256:865d926069c85e6b7b75471756d6d69a11bf9c498927ac90aea12f2f3e81fe9f","target":"graph","created_at":"2026-07-05T11:12:45Z","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/2505.24273/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent breakthroughs in large language models (LLMs) have effectively improved their reasoning abilities, particularly on mathematical and logical problems that have verifiable answers, through techniques such as supervised finetuning (SFT) and reinforcement learning (RL). Prior research indicates that RL effectively internalizes search strategies, enabling long chain-of-thought (CoT) reasoning, with backtracking emerging naturally as a learned capability. However, the precise benefits of backtracking, specifically, how significantly it contributes to reasoning improvements and the optimal ext","authors_text":"Bhuwan Dhingra, Hongyi James Cai, Junlin Wang, Xiaoyin Chen","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-30T06:49:00Z","title":"How Much Backtracking is Enough? Exploring the Interplay of SFT and RL in Enhancing LLM Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24273","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":"sha256:4d187615e9ec9888b062cdc4b3e854abc01b486e75895e9bf609e4f0796ea852","target":"record","created_at":"2026-07-05T11:12:45Z","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":"660e90bf69942a87a964461f11cdf32cfbb3cc1275083f9e90acbc2b51593b29","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-30T06:49:00Z","title_canon_sha256":"4e48f92558ab3814e21a43e6972645e95f1d9e708a2bac053372d078a6d6a3c2"},"schema_version":"1.0","source":{"id":"2505.24273","kind":"arxiv","version":1}},"canonical_sha256":"f2e794261787beaa8d29137c097226928d2f2bdd55d7f38a957aa7a8004a4d47","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f2e794261787beaa8d29137c097226928d2f2bdd55d7f38a957aa7a8004a4d47","first_computed_at":"2026-07-05T11:12:45.696313Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:12:45.696313Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vYbH4g6GgaDMmJQWcj0iVW1IHrdunGZcbkJQKc+ze9kGVwuysfsXPhWm5Ha0fhHb0sc3j/XSdwoJHF7zgynrCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:12:45.696911Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.24273","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4d187615e9ec9888b062cdc4b3e854abc01b486e75895e9bf609e4f0796ea852","sha256:865d926069c85e6b7b75471756d6d69a11bf9c498927ac90aea12f2f3e81fe9f"],"state_sha256":"6c3d29fb4e30cb79d7b4a2cc3c64b2bb7547c29e69bad7c671fe5f93b37c209c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3oNc0QoaZGSxJQ1ZGpM/dt+Y3b4+epyJ7UxnH1RWvWoolxkt7kl6dNyfQ3bWuwPyag4osoZrhU8Z0vqlObrMAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T09:25:53.459490Z","bundle_sha256":"022092509355d4a9eec44ce52f838ea90620419a62e9969ff175881fedd94a30"}}