{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ZTCDUZCNT37NLXFCKRJW7EIMTF","short_pith_number":"pith:ZTCDUZCN","canonical_record":{"source":{"id":"2506.06395","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T19:55:15Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"e203482ea520cc5ee9e4f59decdb773f60846f2380fb2755758da36b6310fa3d","abstract_canon_sha256":"cab533dccb6630d887a7febd84fe95d2ceff093b7a9be65bfbd086a5c46dc92b"},"schema_version":"1.0"},"canonical_sha256":"ccc43a644d9efed5dca254536f910c9946cbe25433731b1d659fd73c0692d2ab","source":{"kind":"arxiv","id":"2506.06395","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.06395","created_at":"2026-07-05T11:19:28Z"},{"alias_kind":"arxiv_version","alias_value":"2506.06395v3","created_at":"2026-07-05T11:19:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.06395","created_at":"2026-07-05T11:19:28Z"},{"alias_kind":"pith_short_12","alias_value":"ZTCDUZCNT37N","created_at":"2026-07-05T11:19:28Z"},{"alias_kind":"pith_short_16","alias_value":"ZTCDUZCNT37NLXFC","created_at":"2026-07-05T11:19:28Z"},{"alias_kind":"pith_short_8","alias_value":"ZTCDUZCN","created_at":"2026-07-05T11:19:28Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ZTCDUZCNT37NLXFCKRJW7EIMTF","target":"record","payload":{"canonical_record":{"source":{"id":"2506.06395","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T19:55:15Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"e203482ea520cc5ee9e4f59decdb773f60846f2380fb2755758da36b6310fa3d","abstract_canon_sha256":"cab533dccb6630d887a7febd84fe95d2ceff093b7a9be65bfbd086a5c46dc92b"},"schema_version":"1.0"},"canonical_sha256":"ccc43a644d9efed5dca254536f910c9946cbe25433731b1d659fd73c0692d2ab","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:19:28.486029Z","signature_b64":"WtwGZCLh69qHM3sHB0ekGDDO53foLYN5rRoOyXYdz0axKMYlftQ/Jp7wqq+5/T/j2FO47Jajb6YelfUJlG4GAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ccc43a644d9efed5dca254536f910c9946cbe25433731b1d659fd73c0692d2ab","last_reissued_at":"2026-07-05T11:19:28.485486Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:19:28.485486Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.06395","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:19:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zk6NIHIRalFspmLD2obXUqm5/Gfm/rC6Ybeq0pe1fdE3ktD9ZKqbH5jIFb4jgLZ+7khH1M7n3jx4Tgn44I8CDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T07:35:22.983474Z"},"content_sha256":"b9e0aead1fe7889d873c713ebe2a7ae38de647e7f9ba9ec208ee39f0282be2db","schema_version":"1.0","event_id":"sha256:b9e0aead1fe7889d873c713ebe2a7ae38de647e7f9ba9ec208ee39f0282be2db"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ZTCDUZCNT37NLXFCKRJW7EIMTF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Confidence Is All You Need: Few-Shot RL Fine-Tuning of Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Alexander Zubrey, Andrey Kuznetsov, Ivan Oseledets, Matvey Skripkin, Pengyi Li","submitted_at":"2025-06-05T19:55:15Z","abstract_excerpt":"Large language models (LLMs) excel at reasoning, yet post-training remains critical for aligning their behavior with task goals. Existing reinforcement learning (RL) methods often depend on costly human annotations or external reward models. We propose Reinforcement Learning via Self-Confidence (RLSC), which uses the model's own confidence as reward signals-eliminating the need for labels, preference models, or reward engineering. Applied to Qwen2.5-Math-7B with only 16 samples per question and 10 or 20 training steps, RLSC improves accuracy by +13.4% on AIME2024, +21.2% on MATH500, +21.7% on "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.06395","kind":"arxiv","version":3},"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/2506.06395/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":"2026-07-05T11:19:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rpkquJuuG4LkLNA/JBtncxY6yqeJsinZBmlweMHNKr22ztDSAwo5B+e+QWLzN9+hp3cco7TLPcdlsglVseeZBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T07:35:22.983974Z"},"content_sha256":"3ab71d914c568b0b8eda1a00f64a9016aa2c246568dd1405e372c7521b7350e4","schema_version":"1.0","event_id":"sha256:3ab71d914c568b0b8eda1a00f64a9016aa2c246568dd1405e372c7521b7350e4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZTCDUZCNT37NLXFCKRJW7EIMTF/bundle.json","state_url":"https://pith.science/pith/ZTCDUZCNT37NLXFCKRJW7EIMTF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZTCDUZCNT37NLXFCKRJW7EIMTF/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":"2026-08-10T07:35:22Z","links":{"resolver":"https://pith.science/pith/ZTCDUZCNT37NLXFCKRJW7EIMTF","bundle":"https://pith.science/pith/ZTCDUZCNT37NLXFCKRJW7EIMTF/bundle.json","state":"https://pith.science/pith/ZTCDUZCNT37NLXFCKRJW7EIMTF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZTCDUZCNT37NLXFCKRJW7EIMTF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ZTCDUZCNT37NLXFCKRJW7EIMTF","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":"cab533dccb6630d887a7febd84fe95d2ceff093b7a9be65bfbd086a5c46dc92b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T19:55:15Z","title_canon_sha256":"e203482ea520cc5ee9e4f59decdb773f60846f2380fb2755758da36b6310fa3d"},"schema_version":"1.0","source":{"id":"2506.06395","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.06395","created_at":"2026-07-05T11:19:28Z"},{"alias_kind":"arxiv_version","alias_value":"2506.06395v3","created_at":"2026-07-05T11:19:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.06395","created_at":"2026-07-05T11:19:28Z"},{"alias_kind":"pith_short_12","alias_value":"ZTCDUZCNT37N","created_at":"2026-07-05T11:19:28Z"},{"alias_kind":"pith_short_16","alias_value":"ZTCDUZCNT37NLXFC","created_at":"2026-07-05T11:19:28Z"},{"alias_kind":"pith_short_8","alias_value":"ZTCDUZCN","created_at":"2026-07-05T11:19:28Z"}],"graph_snapshots":[{"event_id":"sha256:3ab71d914c568b0b8eda1a00f64a9016aa2c246568dd1405e372c7521b7350e4","target":"graph","created_at":"2026-07-05T11:19:28Z","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/2506.06395/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) excel at reasoning, yet post-training remains critical for aligning their behavior with task goals. Existing reinforcement learning (RL) methods often depend on costly human annotations or external reward models. We propose Reinforcement Learning via Self-Confidence (RLSC), which uses the model's own confidence as reward signals-eliminating the need for labels, preference models, or reward engineering. Applied to Qwen2.5-Math-7B with only 16 samples per question and 10 or 20 training steps, RLSC improves accuracy by +13.4% on AIME2024, +21.2% on MATH500, +21.7% on ","authors_text":"Alexander Zubrey, Andrey Kuznetsov, Ivan Oseledets, Matvey Skripkin, Pengyi Li","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T19:55:15Z","title":"Confidence Is All You Need: Few-Shot RL Fine-Tuning of Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.06395","kind":"arxiv","version":3},"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:b9e0aead1fe7889d873c713ebe2a7ae38de647e7f9ba9ec208ee39f0282be2db","target":"record","created_at":"2026-07-05T11:19:28Z","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":"cab533dccb6630d887a7febd84fe95d2ceff093b7a9be65bfbd086a5c46dc92b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T19:55:15Z","title_canon_sha256":"e203482ea520cc5ee9e4f59decdb773f60846f2380fb2755758da36b6310fa3d"},"schema_version":"1.0","source":{"id":"2506.06395","kind":"arxiv","version":3}},"canonical_sha256":"ccc43a644d9efed5dca254536f910c9946cbe25433731b1d659fd73c0692d2ab","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ccc43a644d9efed5dca254536f910c9946cbe25433731b1d659fd73c0692d2ab","first_computed_at":"2026-07-05T11:19:28.485486Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:19:28.485486Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"WtwGZCLh69qHM3sHB0ekGDDO53foLYN5rRoOyXYdz0axKMYlftQ/Jp7wqq+5/T/j2FO47Jajb6YelfUJlG4GAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:19:28.486029Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.06395","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b9e0aead1fe7889d873c713ebe2a7ae38de647e7f9ba9ec208ee39f0282be2db","sha256:3ab71d914c568b0b8eda1a00f64a9016aa2c246568dd1405e372c7521b7350e4"],"state_sha256":"38a739f6ed9a16c133049d1630f3ad08db5e692175639ff51db0f72450bc0fa1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ra5hoOkW2CvdWE5vGsvLlaapNvjex03GliabYOl7lmn1Fhbi89IQDpc2Xo2vrQW8IFurEdVJW3r6N6FdsFTBAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T07:35:22.987346Z","bundle_sha256":"a990a35837bc635fbd6cdc1c5bca3972e0f16b9114f4217483cb0297b6d288c3"}}