{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:UDCHASBKWKWS2KPU6U37XJSCGO","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":"c8e2e66a52a27878649b834cc716c4a6a7d41a6f3f3e6399305e894d67bd7551","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-24T10:54:46Z","title_canon_sha256":"c24f7d718356e4f4cb71afc28037869d67e3aa4ea1762ff1abe07d7227bfaf5d"},"schema_version":"1.0","source":{"id":"2605.24998","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.24998","created_at":"2026-05-26T01:04:09Z"},{"alias_kind":"arxiv_version","alias_value":"2605.24998v1","created_at":"2026-05-26T01:04:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.24998","created_at":"2026-05-26T01:04:09Z"},{"alias_kind":"pith_short_12","alias_value":"UDCHASBKWKWS","created_at":"2026-05-26T01:04:09Z"},{"alias_kind":"pith_short_16","alias_value":"UDCHASBKWKWS2KPU","created_at":"2026-05-26T01:04:09Z"},{"alias_kind":"pith_short_8","alias_value":"UDCHASBK","created_at":"2026-05-26T01:04:09Z"}],"graph_snapshots":[{"event_id":"sha256:c2173a2d19815a2c8907977d07e78241b5c0904abb6f6de1785175cf92bcff76","target":"graph","created_at":"2026-05-26T01:04:09Z","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/2605.24998/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Self-improvement training enables the large reasoning models (LRMs) to improve themselves by self-generating reasoning trajectories as training data without external supervision. However, we find that this method often falls short in complex reasoning tasks and even leads to model collapse. Through a series of preliminary analyses, we reveal two problems: (1) data imbalance, where most training samples are simple, but the challenging yet crucial samples are scarce; (2) overthinking, where many undesired samples with redundant reasoning steps are used for self-training. To this end, we propose ","authors_text":"Bo Du, Dacheng Tao, Juhua Liu, Leszek Rutkowski, Liang Ding, Qihuang Zhong","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-24T10:54:46Z","title":"Better, Faster: Harnessing Self-Improvement in Large Reasoning Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.24998","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":"sha256:3ba119f3e1e98a8f85baaacd57a0c6bb9143641d5a2383ffcce1d3481f0abb12","target":"record","created_at":"2026-05-26T01:04:09Z","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":"c8e2e66a52a27878649b834cc716c4a6a7d41a6f3f3e6399305e894d67bd7551","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-24T10:54:46Z","title_canon_sha256":"c24f7d718356e4f4cb71afc28037869d67e3aa4ea1762ff1abe07d7227bfaf5d"},"schema_version":"1.0","source":{"id":"2605.24998","kind":"arxiv","version":1}},"canonical_sha256":"a0c470482ab2ad2d29f4f537fba64233897083614f134179cfc38ce9e0c48525","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a0c470482ab2ad2d29f4f537fba64233897083614f134179cfc38ce9e0c48525","first_computed_at":"2026-05-26T01:04:09.888821Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-26T01:04:09.888821Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"F2wfsSnqPzL0SUn7/NMzUq7T9tvTn4WkRkfgCS+S4JsE8iLrbHvExyOAneoEWQmbAjVDlYPrTyrUvbjYaenwDw==","signature_status":"signed_v1","signed_at":"2026-05-26T01:04:09.889595Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.24998","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3ba119f3e1e98a8f85baaacd57a0c6bb9143641d5a2383ffcce1d3481f0abb12","sha256:c2173a2d19815a2c8907977d07e78241b5c0904abb6f6de1785175cf92bcff76"],"state_sha256":"6f38b3cd231ae673a0c30a6a8c126c75b6f370cb182fbddd510fd615e4e4274c"}