{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:TFN6GMG3RERLR5O47KFP546AOR","short_pith_number":"pith:TFN6GMG3","canonical_record":{"source":{"id":"2505.21067","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-27T11:52:41Z","cross_cats_sorted":[],"title_canon_sha256":"6efa3267d33641e323019d7a3c88224bb04c38f32e98375a26a9d80fcfeb11e8","abstract_canon_sha256":"a055517c70284e145f3b1a85da326facef6e9db2ad8b15788d9ec13e0cc1823d"},"schema_version":"1.0"},"canonical_sha256":"995be330db8922b8f5dcfa8afef3c0745a91fd2539aaa86cb13fed08d7764295","source":{"kind":"arxiv","id":"2505.21067","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.21067","created_at":"2026-07-05T11:10:23Z"},{"alias_kind":"arxiv_version","alias_value":"2505.21067v1","created_at":"2026-07-05T11:10:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.21067","created_at":"2026-07-05T11:10:23Z"},{"alias_kind":"pith_short_12","alias_value":"TFN6GMG3RERL","created_at":"2026-07-05T11:10:23Z"},{"alias_kind":"pith_short_16","alias_value":"TFN6GMG3RERLR5O4","created_at":"2026-07-05T11:10:23Z"},{"alias_kind":"pith_short_8","alias_value":"TFN6GMG3","created_at":"2026-07-05T11:10:23Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:TFN6GMG3RERLR5O47KFP546AOR","target":"record","payload":{"canonical_record":{"source":{"id":"2505.21067","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-27T11:52:41Z","cross_cats_sorted":[],"title_canon_sha256":"6efa3267d33641e323019d7a3c88224bb04c38f32e98375a26a9d80fcfeb11e8","abstract_canon_sha256":"a055517c70284e145f3b1a85da326facef6e9db2ad8b15788d9ec13e0cc1823d"},"schema_version":"1.0"},"canonical_sha256":"995be330db8922b8f5dcfa8afef3c0745a91fd2539aaa86cb13fed08d7764295","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:23.134144Z","signature_b64":"Htga0p2iXKP3BTg86XFJgmiDV3OhmTTQF+tzpOJlIvNMHjznukTUpehiw3fk8r+SVB7FebXkhBsGh4HuCJeECQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"995be330db8922b8f5dcfa8afef3c0745a91fd2539aaa86cb13fed08d7764295","last_reissued_at":"2026-07-05T11:10:23.133635Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:23.133635Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.21067","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":"2026-07-05T11:10:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qDH4wYSWOofidITySa1YvBO/CtRFq+FbB88IF13/md89Yevz1q9Y0h+UlG/EkMpK8Tb9TmalxmSYDLYAboc9BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-25T12:31:29.109838Z"},"content_sha256":"6e5976e263b3946bdf4fa7fff9483e9b0a0116c3a46078514cdb0b1082389ed8","schema_version":"1.0","event_id":"sha256:6e5976e263b3946bdf4fa7fff9483e9b0a0116c3a46078514cdb0b1082389ed8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:TFN6GMG3RERLR5O47KFP546AOR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Why Distillation can Outperform Zero-RL: The Role of Flexible Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bin Wen, Fan Yang, Guorui Zhou, Liyuan Mao, Tianke Zhang, Tingting Gao, Xiao Hu, Xingyu Lu, Yifan Zhang","submitted_at":"2025-05-27T11:52:41Z","abstract_excerpt":"Reinforcement learning (RL) has played an important role in improving the reasoning ability of large language models (LLMs). Some studies apply RL directly to \\textit{smaller} base models (known as zero-RL) and also achieve notable progress. However, in this paper, we show that using only 920 examples, a simple distillation method based on the base model can clearly outperform zero-RL, which typically requires much more data and computational cost. By analyzing the token frequency in model outputs, we find that the distilled model shows more flexible reasoning. It uses anthropomorphic tokens a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.21067","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/2505.21067/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":"2026-07-05T11:10:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Yd+zNSYHAr6q8x6Daxkbp9YMqAVPuDOAXuIXQR9pzWUL1bs9B3Ek53zixKmO7S6zR8NHKnhS/5B+wBTXNKnJDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-25T12:31:29.110219Z"},"content_sha256":"208918f6524eeb172e0c1101d02a997047dca80698f10aca4525094d8c369899","schema_version":"1.0","event_id":"sha256:208918f6524eeb172e0c1101d02a997047dca80698f10aca4525094d8c369899"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TFN6GMG3RERLR5O47KFP546AOR/bundle.json","state_url":"https://pith.science/pith/TFN6GMG3RERLR5O47KFP546AOR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TFN6GMG3RERLR5O47KFP546AOR/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":"2026-07-25T12:31:29Z","links":{"resolver":"https://pith.science/pith/TFN6GMG3RERLR5O47KFP546AOR","bundle":"https://pith.science/pith/TFN6GMG3RERLR5O47KFP546AOR/bundle.json","state":"https://pith.science/pith/TFN6GMG3RERLR5O47KFP546AOR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TFN6GMG3RERLR5O47KFP546AOR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:TFN6GMG3RERLR5O47KFP546AOR","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":"a055517c70284e145f3b1a85da326facef6e9db2ad8b15788d9ec13e0cc1823d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-27T11:52:41Z","title_canon_sha256":"6efa3267d33641e323019d7a3c88224bb04c38f32e98375a26a9d80fcfeb11e8"},"schema_version":"1.0","source":{"id":"2505.21067","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.21067","created_at":"2026-07-05T11:10:23Z"},{"alias_kind":"arxiv_version","alias_value":"2505.21067v1","created_at":"2026-07-05T11:10:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.21067","created_at":"2026-07-05T11:10:23Z"},{"alias_kind":"pith_short_12","alias_value":"TFN6GMG3RERL","created_at":"2026-07-05T11:10:23Z"},{"alias_kind":"pith_short_16","alias_value":"TFN6GMG3RERLR5O4","created_at":"2026-07-05T11:10:23Z"},{"alias_kind":"pith_short_8","alias_value":"TFN6GMG3","created_at":"2026-07-05T11:10:23Z"}],"graph_snapshots":[{"event_id":"sha256:208918f6524eeb172e0c1101d02a997047dca80698f10aca4525094d8c369899","target":"graph","created_at":"2026-07-05T11:10:23Z","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/2505.21067/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning (RL) has played an important role in improving the reasoning ability of large language models (LLMs). Some studies apply RL directly to \\textit{smaller} base models (known as zero-RL) and also achieve notable progress. However, in this paper, we show that using only 920 examples, a simple distillation method based on the base model can clearly outperform zero-RL, which typically requires much more data and computational cost. By analyzing the token frequency in model outputs, we find that the distilled model shows more flexible reasoning. It uses anthropomorphic tokens a","authors_text":"Bin Wen, Fan Yang, Guorui Zhou, Liyuan Mao, Tianke Zhang, Tingting Gao, Xiao Hu, Xingyu Lu, Yifan Zhang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-27T11:52:41Z","title":"Why Distillation can Outperform Zero-RL: The Role of Flexible Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.21067","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":"sha256:6e5976e263b3946bdf4fa7fff9483e9b0a0116c3a46078514cdb0b1082389ed8","target":"record","created_at":"2026-07-05T11:10:23Z","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":"a055517c70284e145f3b1a85da326facef6e9db2ad8b15788d9ec13e0cc1823d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-27T11:52:41Z","title_canon_sha256":"6efa3267d33641e323019d7a3c88224bb04c38f32e98375a26a9d80fcfeb11e8"},"schema_version":"1.0","source":{"id":"2505.21067","kind":"arxiv","version":1}},"canonical_sha256":"995be330db8922b8f5dcfa8afef3c0745a91fd2539aaa86cb13fed08d7764295","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"995be330db8922b8f5dcfa8afef3c0745a91fd2539aaa86cb13fed08d7764295","first_computed_at":"2026-07-05T11:10:23.133635Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:10:23.133635Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Htga0p2iXKP3BTg86XFJgmiDV3OhmTTQF+tzpOJlIvNMHjznukTUpehiw3fk8r+SVB7FebXkhBsGh4HuCJeECQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:10:23.134144Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.21067","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6e5976e263b3946bdf4fa7fff9483e9b0a0116c3a46078514cdb0b1082389ed8","sha256:208918f6524eeb172e0c1101d02a997047dca80698f10aca4525094d8c369899"],"state_sha256":"7012e7e08c5cf3415fc719c07192810af2e1174a50bc738ff6067f2735b4e3be"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"at6ckL1Pnz1r04FnXt6/btupogF5S1JvokThWdwbc9BmFHq7CCvY8EXxCG8EVUAGVTN0FEznLw4L/53MN1UFDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-25T12:31:29.113576Z","bundle_sha256":"9ef6d6658b6d5dab949ffb5a56da0e3a88a765186b23f9d0324dfd40a7819a6c"}}