{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:PAK2NUIABZI2AWNA2VWHUGYKUL","short_pith_number":"pith:PAK2NUIA","schema_version":"1.0","canonical_sha256":"7815a6d1000e51a059a0d56c7a1b0aa2ea64bc9151b67e3e64bd2f69f0deedcc","source":{"kind":"arxiv","id":"2605.15454","version":1},"attestation_state":"computed","paper":{"title":"Reasoning Models Don't Just Think Longer, They Move Differently","license":"http://creativecommons.org/licenses/by/4.0/","headline":"After length correction, reasoning models show distinct hidden-state trajectories on harder problems.","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Anders Gj{\\o}lbye, Lars Kai Hansen, Sanmi Koyejo","submitted_at":"2026-05-14T22:37:33Z","abstract_excerpt":"Reasoning-trained language models often spend more tokens on harder problems, but longer chains of thought do not show whether a model is merely computing for more steps or following a different internal trajectory. We study this distinction through hidden-state trajectories during chain-of-thought generation across competitive programming, mathematics, and Boolean satisfiability. Raw trajectory geometry is strongly shaped by generation length: longer generations mechanically alter path statistics, so difficulty-dependent comparisons are misleading without adjustment. After residualizing traje"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.15454","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-14T22:37:33Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"2ee461029fdeeeebb789b1d76858956f16d334d95a202ce589669959549b5ff3","abstract_canon_sha256":"6987374dc451519ff851919f375b21ea8a49bf936b0529d23c427e96cd1aa356"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:00:59.438454Z","signature_b64":"nMElWnR+8caqGNuicC2TF+S9CdGY+E3Cz3ipxSLG4ar+xD7udMKKriB31R/EwqY2XmYgSBpR+GEqnZwZlo5jAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7815a6d1000e51a059a0d56c7a1b0aa2ea64bc9151b67e3e64bd2f69f0deedcc","last_reissued_at":"2026-05-20T00:00:59.437691Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:00:59.437691Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reasoning Models Don't Just Think Longer, They Move Differently","license":"http://creativecommons.org/licenses/by/4.0/","headline":"After length correction, reasoning models show distinct hidden-state trajectories on harder problems.","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Anders Gj{\\o}lbye, Lars Kai Hansen, Sanmi Koyejo","submitted_at":"2026-05-14T22:37:33Z","abstract_excerpt":"Reasoning-trained language models often spend more tokens on harder problems, but longer chains of thought do not show whether a model is merely computing for more steps or following a different internal trajectory. We study this distinction through hidden-state trajectories during chain-of-thought generation across competitive programming, mathematics, and Boolean satisfiability. Raw trajectory geometry is strongly shaped by generation length: longer generations mechanically alter path statistics, so difficulty-dependent comparisons are misleading without adjustment. After residualizing traje"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"After residualizing trajectory statistics on length, difficulty remains systematically coupled to corrected trajectory geometry across all domains studied. The clearest reasoning-specific separation appears in the code domain, where harder problems show more direct corrected trajectories and less heterogeneous local curvature in reasoning-trained models than in matched instruction-tuned baselines.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That residualizing raw trajectory statistics on generation length successfully removes all mechanical length-induced effects and leaves only the component attributable to reasoning strategy or internal computation.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"After length-correcting hidden-state trajectories during chain-of-thought, reasoning models show systematically different geometry on harder problems than baselines, strongest in competitive programming.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"After length correction, reasoning models show distinct hidden-state trajectories on harder problems.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"85f9268695eb2cfd6131799219a41f072b599757cd8b2ec8981df2744cae394f"},"source":{"id":"2605.15454","kind":"arxiv","version":1},"verdict":{"id":"7b502532-69c7-4d89-bdda-a77467dddae6","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T14:39:02.963239Z","strongest_claim":"After residualizing trajectory statistics on length, difficulty remains systematically coupled to corrected trajectory geometry across all domains studied. The clearest reasoning-specific separation appears in the code domain, where harder problems show more direct corrected trajectories and less heterogeneous local curvature in reasoning-trained models than in matched instruction-tuned baselines.","one_line_summary":"After length-correcting hidden-state trajectories during chain-of-thought, reasoning models show systematically different geometry on harder problems than baselines, strongest in competitive programming.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That residualizing raw trajectory statistics on generation length successfully removes all mechanical length-induced effects and leaves only the component attributable to reasoning strategy or internal computation.","pith_extraction_headline":"After length correction, reasoning models show distinct hidden-state trajectories on harder problems."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.15454/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"cited_work_retraction","ran_at":"2026-05-19T15:51:55.289286Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"citation_quote_validity","ran_at":"2026-05-19T15:49:52.204121Z","status":"completed","version":"0.1.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T15:01:17.612094Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T14:50:17.313748Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T14:21:54.106172Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T13:33:22.675294Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"0f17cfab8e6dac1d12ecc3804fe9e585399e81eacf7ed7ab52ad8cdf7f746b79"},"references":{"count":14,"sample":[{"doi":"","year":null,"title":"Mitigating overthinking in large reasoning models via manifold steering","work_id":"f8c2a7ce-0438-463d-8985-cfd021da58cb","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2065,"title":"Jolliffe and Jorge Cadima","work_id":"5b863ffc-2076-49ed-889b-81ecc3cf2f65","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Probing the difficulty perception mechanism of large language models","work_id":"b131cbd5-0794-4a78-a1db-5d7f274ea2fb","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"LLMs encode how difficult problems are.arXiv preprint arXiv:2510.18147,","work_id":"6948d73b-e278-4096-8f01-0856a18ca1ba","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters","work_id":"a8d50b24-bdf5-46ed-bc4f-2927dfd81f1d","ref_index":5,"cited_arxiv_id":"2408.03314","is_internal_anchor":true}],"resolved_work":14,"snapshot_sha256":"d8e12ec78efbfd91823462f19e54c49c88ddef77fd875e4bb3bf612409b65a30","internal_anchors":2},"formal_canon":{"evidence_count":2,"snapshot_sha256":"a92230f44cfaf3190036fd61fcc4d094ec508df946b5ca722110c1747436e564"},"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":"2605.15454","created_at":"2026-05-20T00:00:59.437807+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.15454v1","created_at":"2026-05-20T00:00:59.437807+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15454","created_at":"2026-05-20T00:00:59.437807+00:00"},{"alias_kind":"pith_short_12","alias_value":"PAK2NUIABZI2","created_at":"2026-05-20T00:00:59.437807+00:00"},{"alias_kind":"pith_short_16","alias_value":"PAK2NUIABZI2AWNA","created_at":"2026-05-20T00:00:59.437807+00:00"},{"alias_kind":"pith_short_8","alias_value":"PAK2NUIA","created_at":"2026-05-20T00:00:59.437807+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PAK2NUIABZI2AWNA2VWHUGYKUL","json":"https://pith.science/pith/PAK2NUIABZI2AWNA2VWHUGYKUL.json","graph_json":"https://pith.science/api/pith-number/PAK2NUIABZI2AWNA2VWHUGYKUL/graph.json","events_json":"https://pith.science/api/pith-number/PAK2NUIABZI2AWNA2VWHUGYKUL/events.json","paper":"https://pith.science/paper/PAK2NUIA"},"agent_actions":{"view_html":"https://pith.science/pith/PAK2NUIABZI2AWNA2VWHUGYKUL","download_json":"https://pith.science/pith/PAK2NUIABZI2AWNA2VWHUGYKUL.json","view_paper":"https://pith.science/paper/PAK2NUIA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.15454&json=true","fetch_graph":"https://pith.science/api/pith-number/PAK2NUIABZI2AWNA2VWHUGYKUL/graph.json","fetch_events":"https://pith.science/api/pith-number/PAK2NUIABZI2AWNA2VWHUGYKUL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PAK2NUIABZI2AWNA2VWHUGYKUL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PAK2NUIABZI2AWNA2VWHUGYKUL/action/storage_attestation","attest_author":"https://pith.science/pith/PAK2NUIABZI2AWNA2VWHUGYKUL/action/author_attestation","sign_citation":"https://pith.science/pith/PAK2NUIABZI2AWNA2VWHUGYKUL/action/citation_signature","submit_replication":"https://pith.science/pith/PAK2NUIABZI2AWNA2VWHUGYKUL/action/replication_record"}},"created_at":"2026-05-20T00:00:59.437807+00:00","updated_at":"2026-05-20T00:00:59.437807+00:00"}