{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:5XH4K5R6ZDBYMAGAREWH4IGOUU","short_pith_number":"pith:5XH4K5R6","schema_version":"1.0","canonical_sha256":"edcfc5763ec8c38600c0892c7e20cea516cb0244e09b63c6b5ea319701c2bc06","source":{"kind":"arxiv","id":"2606.05464","version":1},"attestation_state":"computed","paper":{"title":"Step-by-Step Optimization-like Reasoning in LLMs over Expanding Search Spaces","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Mihaela van der Schaar, Nabeel Seedat, Nicol\\'as Astorga","submitted_at":"2026-06-03T21:43:38Z","abstract_excerpt":"Verifiable reward training has improved mathematical and coding reasoning, but these domains capture only part of step-by-step decision making. Many real-world tasks require finding a high-value feasible plan among many valid alternatives. We introduce OPT*, a scalable family of optimization-style tasks for training and evaluating LLM step-by-step optimization-like reasoning along a complexity axis: each task provides a feasibility checker and evaluator, while a complexity parameter expands the search space without requiring new human labels. This motivates studying these tasks in two regimes:"},"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":"2606.05464","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-03T21:43:38Z","cross_cats_sorted":[],"title_canon_sha256":"2e5546527f6fedbf680a4d7f35b6c472ae1e70d55bf26ec7da3ac2fe5880fe7c","abstract_canon_sha256":"b8a395204a550a58a854b84219426030bf9bcdcaf4ef7dd52aa9817cacbbdcd8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-05T01:14:51.939585Z","signature_b64":"jNOXE6fKdrBFcr1xr0T+q9xurDCEztG25fcdrgUSP83ZWnxGvxQWs/Wl72fyq+0XJYDFc+1uNVmyJrB3Rq4JBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"edcfc5763ec8c38600c0892c7e20cea516cb0244e09b63c6b5ea319701c2bc06","last_reissued_at":"2026-06-05T01:14:51.939086Z","signature_status":"signed_v1","first_computed_at":"2026-06-05T01:14:51.939086Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Step-by-Step Optimization-like Reasoning in LLMs over Expanding Search Spaces","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Mihaela van der Schaar, Nabeel Seedat, Nicol\\'as Astorga","submitted_at":"2026-06-03T21:43:38Z","abstract_excerpt":"Verifiable reward training has improved mathematical and coding reasoning, but these domains capture only part of step-by-step decision making. Many real-world tasks require finding a high-value feasible plan among many valid alternatives. We introduce OPT*, a scalable family of optimization-style tasks for training and evaluating LLM step-by-step optimization-like reasoning along a complexity axis: each task provides a feasibility checker and evaluator, while a complexity parameter expands the search space without requiring new human labels. This motivates studying these tasks in two regimes:"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.05464","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/2606.05464/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":"2606.05464","created_at":"2026-06-05T01:14:51.939146+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.05464v1","created_at":"2026-06-05T01:14:51.939146+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.05464","created_at":"2026-06-05T01:14:51.939146+00:00"},{"alias_kind":"pith_short_12","alias_value":"5XH4K5R6ZDBY","created_at":"2026-06-05T01:14:51.939146+00:00"},{"alias_kind":"pith_short_16","alias_value":"5XH4K5R6ZDBYMAGA","created_at":"2026-06-05T01:14:51.939146+00:00"},{"alias_kind":"pith_short_8","alias_value":"5XH4K5R6","created_at":"2026-06-05T01:14:51.939146+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/5XH4K5R6ZDBYMAGAREWH4IGOUU","json":"https://pith.science/pith/5XH4K5R6ZDBYMAGAREWH4IGOUU.json","graph_json":"https://pith.science/api/pith-number/5XH4K5R6ZDBYMAGAREWH4IGOUU/graph.json","events_json":"https://pith.science/api/pith-number/5XH4K5R6ZDBYMAGAREWH4IGOUU/events.json","paper":"https://pith.science/paper/5XH4K5R6"},"agent_actions":{"view_html":"https://pith.science/pith/5XH4K5R6ZDBYMAGAREWH4IGOUU","download_json":"https://pith.science/pith/5XH4K5R6ZDBYMAGAREWH4IGOUU.json","view_paper":"https://pith.science/paper/5XH4K5R6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.05464&json=true","fetch_graph":"https://pith.science/api/pith-number/5XH4K5R6ZDBYMAGAREWH4IGOUU/graph.json","fetch_events":"https://pith.science/api/pith-number/5XH4K5R6ZDBYMAGAREWH4IGOUU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5XH4K5R6ZDBYMAGAREWH4IGOUU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5XH4K5R6ZDBYMAGAREWH4IGOUU/action/storage_attestation","attest_author":"https://pith.science/pith/5XH4K5R6ZDBYMAGAREWH4IGOUU/action/author_attestation","sign_citation":"https://pith.science/pith/5XH4K5R6ZDBYMAGAREWH4IGOUU/action/citation_signature","submit_replication":"https://pith.science/pith/5XH4K5R6ZDBYMAGAREWH4IGOUU/action/replication_record"}},"created_at":"2026-06-05T01:14:51.939146+00:00","updated_at":"2026-06-05T01:14:51.939146+00:00"}