{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:XXUUTPXDV4FHY4A5XOFTJDAUO3","short_pith_number":"pith:XXUUTPXD","schema_version":"1.0","canonical_sha256":"bde949bee3af0a7c701dbb8b348c1476ed2bdb751ca016a7c92a2537d3ca1937","source":{"kind":"arxiv","id":"2607.19523","version":1},"attestation_state":"computed","paper":{"title":"When Reasoning Narrows the Move: Diversity Collapse in LLM Game Play","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"David Simchi-Levi, Junyi Sha, Renfei Tan","submitted_at":"2026-07-21T19:10:38Z","abstract_excerpt":"Supervised fine-tuning (SFT) is widely used to adapt large language models to downstream tasks, but its effect on behavioral diversity in sequential decision-making remains under-explored. We study this question in a controlled suite of deterministic board games based on tic-tac-toe variants, where optimal actions are exactly computable and diversity can be measured directly. Across state-level evaluation, arena gameplay, and training trajectories, we find that reasoning-mode generation frequently suppresses action diversity without uniformly improving action accuracy. Furthermore, standard SF"},"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":"2607.19523","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-21T19:10:38Z","cross_cats_sorted":[],"title_canon_sha256":"c505abf968743c1bab91d271080812120ca2e0b819e30385599134a127192b15","abstract_canon_sha256":"56c2f60dcc1b13c6459f20bd8e289abbc3b3ce94c12c9db1713b0ac5fa227363"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-23T00:23:54.974458Z","signature_b64":"09li+eR/BXMKOHS0h8R3jSB7fuK7HcrLVMhuJI+gYOMlzWqGupwMj3MNOB00DTFLJtVvzvcPqSBZl3ndJgc/Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bde949bee3af0a7c701dbb8b348c1476ed2bdb751ca016a7c92a2537d3ca1937","last_reissued_at":"2026-07-23T00:23:54.973640Z","signature_status":"signed_v1","first_computed_at":"2026-07-23T00:23:54.973640Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When Reasoning Narrows the Move: Diversity Collapse in LLM Game Play","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"David Simchi-Levi, Junyi Sha, Renfei Tan","submitted_at":"2026-07-21T19:10:38Z","abstract_excerpt":"Supervised fine-tuning (SFT) is widely used to adapt large language models to downstream tasks, but its effect on behavioral diversity in sequential decision-making remains under-explored. We study this question in a controlled suite of deterministic board games based on tic-tac-toe variants, where optimal actions are exactly computable and diversity can be measured directly. Across state-level evaluation, arena gameplay, and training trajectories, we find that reasoning-mode generation frequently suppresses action diversity without uniformly improving action accuracy. Furthermore, standard SF"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.19523","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/2607.19523/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":"2607.19523","created_at":"2026-07-23T00:23:54.974050+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.19523v1","created_at":"2026-07-23T00:23:54.974050+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.19523","created_at":"2026-07-23T00:23:54.974050+00:00"},{"alias_kind":"pith_short_12","alias_value":"XXUUTPXDV4FH","created_at":"2026-07-23T00:23:54.974050+00:00"},{"alias_kind":"pith_short_16","alias_value":"XXUUTPXDV4FHY4A5","created_at":"2026-07-23T00:23:54.974050+00:00"},{"alias_kind":"pith_short_8","alias_value":"XXUUTPXD","created_at":"2026-07-23T00:23:54.974050+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XXUUTPXDV4FHY4A5XOFTJDAUO3","json":"https://pith.science/pith/XXUUTPXDV4FHY4A5XOFTJDAUO3.json","graph_json":"https://pith.science/api/pith-number/XXUUTPXDV4FHY4A5XOFTJDAUO3/graph.json","events_json":"https://pith.science/api/pith-number/XXUUTPXDV4FHY4A5XOFTJDAUO3/events.json","paper":"https://pith.science/paper/XXUUTPXD"},"agent_actions":{"view_html":"https://pith.science/pith/XXUUTPXDV4FHY4A5XOFTJDAUO3","download_json":"https://pith.science/pith/XXUUTPXDV4FHY4A5XOFTJDAUO3.json","view_paper":"https://pith.science/paper/XXUUTPXD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.19523&json=true","fetch_graph":"https://pith.science/api/pith-number/XXUUTPXDV4FHY4A5XOFTJDAUO3/graph.json","fetch_events":"https://pith.science/api/pith-number/XXUUTPXDV4FHY4A5XOFTJDAUO3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XXUUTPXDV4FHY4A5XOFTJDAUO3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XXUUTPXDV4FHY4A5XOFTJDAUO3/action/storage_attestation","attest_author":"https://pith.science/pith/XXUUTPXDV4FHY4A5XOFTJDAUO3/action/author_attestation","sign_citation":"https://pith.science/pith/XXUUTPXDV4FHY4A5XOFTJDAUO3/action/citation_signature","submit_replication":"https://pith.science/pith/XXUUTPXDV4FHY4A5XOFTJDAUO3/action/replication_record"}},"created_at":"2026-07-23T00:23:54.974050+00:00","updated_at":"2026-07-23T00:23:54.974050+00:00"}