{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YXYJDKPHDOX6ET45LXPTYUEUDW","short_pith_number":"pith:YXYJDKPH","schema_version":"1.0","canonical_sha256":"c5f091a9e71bafe24f9d5ddf3c50941d8a11ffad0f398e2b95267cb94e18513b","source":{"kind":"arxiv","id":"2509.04027","version":3},"attestation_state":"computed","paper":{"title":"CoT-Space: A Theoretical Framework for Internal Slow-Thinking via Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Hao Yi, Yong Liu, Zeyu Gan","submitted_at":"2025-09-04T09:02:16Z","abstract_excerpt":"Test-time scaling, primarily manifested through multi-step Chain-of-Thought (CoT) reasoning via Reinforcement Learning (RL), has emerged as a pivotal paradigm for enhancing the reasoning capabilities of Large Language Models (LLMs). However, a significant theoretical gap persists: traditional token-level analysis fails to capture the macroscopic dynamics of reasoning-level scaling. To address this, we introduce CoT-Space, a novel theoretical framework that recasts the reasoning process from a discrete token-prediction task to an optimization process within a continuous, reasoning-level semanti"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2509.04027","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-09-04T09:02:16Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"5570c26022e798cd0ff70ec00fab394f71af783a5fd6ff94dd9280c277ccb5f6","abstract_canon_sha256":"4a827f80132386c8f35b0c7ccb5e025306a65da28d292cbf7c8dc9e178a958e8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-05T01:15:15.814377Z","signature_b64":"pDnbBQRgBbJVRH8U4FG4E9/QqMSSW256GfpomnU9VMLUv+FMDaA68vtFKKP7KYfZ2O8+aqzermkBExQ5e/a6CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c5f091a9e71bafe24f9d5ddf3c50941d8a11ffad0f398e2b95267cb94e18513b","last_reissued_at":"2026-06-05T01:15:15.813405Z","signature_status":"signed_v1","first_computed_at":"2026-06-05T01:15:15.813405Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CoT-Space: A Theoretical Framework for Internal Slow-Thinking via Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Hao Yi, Yong Liu, Zeyu Gan","submitted_at":"2025-09-04T09:02:16Z","abstract_excerpt":"Test-time scaling, primarily manifested through multi-step Chain-of-Thought (CoT) reasoning via Reinforcement Learning (RL), has emerged as a pivotal paradigm for enhancing the reasoning capabilities of Large Language Models (LLMs). However, a significant theoretical gap persists: traditional token-level analysis fails to capture the macroscopic dynamics of reasoning-level scaling. To address this, we introduce CoT-Space, a novel theoretical framework that recasts the reasoning process from a discrete token-prediction task to an optimization process within a continuous, reasoning-level semanti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.04027","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/2509.04027/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2509.04027","created_at":"2026-06-05T01:15:15.813530+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.04027v3","created_at":"2026-06-05T01:15:15.813530+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.04027","created_at":"2026-06-05T01:15:15.813530+00:00"},{"alias_kind":"pith_short_12","alias_value":"YXYJDKPHDOX6","created_at":"2026-06-05T01:15:15.813530+00:00"},{"alias_kind":"pith_short_16","alias_value":"YXYJDKPHDOX6ET45","created_at":"2026-06-05T01:15:15.813530+00:00"},{"alias_kind":"pith_short_8","alias_value":"YXYJDKPH","created_at":"2026-06-05T01:15:15.813530+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.08827","citing_title":"A Survey of Reinforcement Learning for Large Reasoning Models","ref_index":152,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YXYJDKPHDOX6ET45LXPTYUEUDW","json":"https://pith.science/pith/YXYJDKPHDOX6ET45LXPTYUEUDW.json","graph_json":"https://pith.science/api/pith-number/YXYJDKPHDOX6ET45LXPTYUEUDW/graph.json","events_json":"https://pith.science/api/pith-number/YXYJDKPHDOX6ET45LXPTYUEUDW/events.json","paper":"https://pith.science/paper/YXYJDKPH"},"agent_actions":{"view_html":"https://pith.science/pith/YXYJDKPHDOX6ET45LXPTYUEUDW","download_json":"https://pith.science/pith/YXYJDKPHDOX6ET45LXPTYUEUDW.json","view_paper":"https://pith.science/paper/YXYJDKPH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.04027&json=true","fetch_graph":"https://pith.science/api/pith-number/YXYJDKPHDOX6ET45LXPTYUEUDW/graph.json","fetch_events":"https://pith.science/api/pith-number/YXYJDKPHDOX6ET45LXPTYUEUDW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YXYJDKPHDOX6ET45LXPTYUEUDW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YXYJDKPHDOX6ET45LXPTYUEUDW/action/storage_attestation","attest_author":"https://pith.science/pith/YXYJDKPHDOX6ET45LXPTYUEUDW/action/author_attestation","sign_citation":"https://pith.science/pith/YXYJDKPHDOX6ET45LXPTYUEUDW/action/citation_signature","submit_replication":"https://pith.science/pith/YXYJDKPHDOX6ET45LXPTYUEUDW/action/replication_record"}},"created_at":"2026-06-05T01:15:15.813530+00:00","updated_at":"2026-06-05T01:15:15.813530+00:00"}