{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6WWWHJ4WGQHTVZTHPCZYKEY4UF","short_pith_number":"pith:6WWWHJ4W","schema_version":"1.0","canonical_sha256":"f5ad63a796340f3ae66778b385131ca1480c35823d8af4501f663ddf52e770b8","source":{"kind":"arxiv","id":"2605.29687","version":1},"attestation_state":"computed","paper":{"title":"Reliable Reasoning with Large Language Models via Preference-Based Maximum Satisfiability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LO"],"primary_cat":"cs.AI","authors_text":"Felip Many\\`a, Guillem Aleny\\`a, Marta Kwiatkowska, Pedro Orvalho","submitted_at":"2026-05-28T09:51:33Z","abstract_excerpt":"Large Language Models (LLMs) excel at understanding natural language but struggle with optimisation tasks involving multiple constraints and user-defined preferences, which commonly arise in domains such as robotics. We propose a hybrid reasoning approach in which LLMs externalise reasoning through code generation. Given a natural language problem description, an LLM generates Python code that encodes user-defined constraints and preferences as a preference-based Maximum Satisfiability (MaxSAT) problem, which is then solved by an exact MaxSAT solver. To ensure correctness, solutions returned b"},"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":"2605.29687","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-28T09:51:33Z","cross_cats_sorted":["cs.LO"],"title_canon_sha256":"21eb00e7a3e983d76493c9ce68378dd278e0eb290fdca560f4a0af4666def45e","abstract_canon_sha256":"946711563f52fe469c06655babda8b36fb53d1e7cf6d95cc344882961433dd1a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:55.649086Z","signature_b64":"i+zfiHcVu1BPNOwCtuaAR9ydQFH90Qha3VETcWiXe3IcsdpxWbVW7Hfkk9d9QFtJTsOvC9WlP4ODEQVg6SqpCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f5ad63a796340f3ae66778b385131ca1480c35823d8af4501f663ddf52e770b8","last_reissued_at":"2026-05-29T01:05:55.648523Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:55.648523Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reliable Reasoning with Large Language Models via Preference-Based Maximum Satisfiability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LO"],"primary_cat":"cs.AI","authors_text":"Felip Many\\`a, Guillem Aleny\\`a, Marta Kwiatkowska, Pedro Orvalho","submitted_at":"2026-05-28T09:51:33Z","abstract_excerpt":"Large Language Models (LLMs) excel at understanding natural language but struggle with optimisation tasks involving multiple constraints and user-defined preferences, which commonly arise in domains such as robotics. We propose a hybrid reasoning approach in which LLMs externalise reasoning through code generation. Given a natural language problem description, an LLM generates Python code that encodes user-defined constraints and preferences as a preference-based Maximum Satisfiability (MaxSAT) problem, which is then solved by an exact MaxSAT solver. To ensure correctness, solutions returned b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29687","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/2605.29687/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":"2605.29687","created_at":"2026-05-29T01:05:55.648603+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.29687v1","created_at":"2026-05-29T01:05:55.648603+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29687","created_at":"2026-05-29T01:05:55.648603+00:00"},{"alias_kind":"pith_short_12","alias_value":"6WWWHJ4WGQHT","created_at":"2026-05-29T01:05:55.648603+00:00"},{"alias_kind":"pith_short_16","alias_value":"6WWWHJ4WGQHTVZTH","created_at":"2026-05-29T01:05:55.648603+00:00"},{"alias_kind":"pith_short_8","alias_value":"6WWWHJ4W","created_at":"2026-05-29T01:05:55.648603+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/6WWWHJ4WGQHTVZTHPCZYKEY4UF","json":"https://pith.science/pith/6WWWHJ4WGQHTVZTHPCZYKEY4UF.json","graph_json":"https://pith.science/api/pith-number/6WWWHJ4WGQHTVZTHPCZYKEY4UF/graph.json","events_json":"https://pith.science/api/pith-number/6WWWHJ4WGQHTVZTHPCZYKEY4UF/events.json","paper":"https://pith.science/paper/6WWWHJ4W"},"agent_actions":{"view_html":"https://pith.science/pith/6WWWHJ4WGQHTVZTHPCZYKEY4UF","download_json":"https://pith.science/pith/6WWWHJ4WGQHTVZTHPCZYKEY4UF.json","view_paper":"https://pith.science/paper/6WWWHJ4W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.29687&json=true","fetch_graph":"https://pith.science/api/pith-number/6WWWHJ4WGQHTVZTHPCZYKEY4UF/graph.json","fetch_events":"https://pith.science/api/pith-number/6WWWHJ4WGQHTVZTHPCZYKEY4UF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6WWWHJ4WGQHTVZTHPCZYKEY4UF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6WWWHJ4WGQHTVZTHPCZYKEY4UF/action/storage_attestation","attest_author":"https://pith.science/pith/6WWWHJ4WGQHTVZTHPCZYKEY4UF/action/author_attestation","sign_citation":"https://pith.science/pith/6WWWHJ4WGQHTVZTHPCZYKEY4UF/action/citation_signature","submit_replication":"https://pith.science/pith/6WWWHJ4WGQHTVZTHPCZYKEY4UF/action/replication_record"}},"created_at":"2026-05-29T01:05:55.648603+00:00","updated_at":"2026-05-29T01:05:55.648603+00:00"}