{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YNJOHVF2S7CJRYO5U4CBWH7YBZ","short_pith_number":"pith:YNJOHVF2","schema_version":"1.0","canonical_sha256":"c352e3d4ba97c498e1dda7041b1ff80e568000a2521d4ff5c66bf985c68b5b31","source":{"kind":"arxiv","id":"2508.20907","version":1},"attestation_state":"computed","paper":{"title":"Quantum Verifiable Rewards for Post-Training Qiskit Code Assistant","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"quant-ph","authors_text":"Adarsh Tiwari, David Kremer, Ismael Faro, Juan Cruz-Benito, Nicolas Dupuis, Youssef Mroueh","submitted_at":"2025-08-28T15:37:40Z","abstract_excerpt":"Qiskit is an open-source quantum computing framework that allows users to design, simulate, and run quantum circuits on real quantum hardware. We explore post-training techniques for LLMs to assist in writing Qiskit code. We introduce quantum verification as an effective method for ensuring code quality and executability on quantum hardware. To support this, we developed a synthetic data pipeline that generates quantum problem-unit test pairs and used it to create preference data for aligning LLMs with DPO. Additionally, we trained models using GRPO, leveraging quantum-verifiable rewards provi"},"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":"2508.20907","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"quant-ph","submitted_at":"2025-08-28T15:37:40Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6587eadd95cd4f92be322ed0bf3bbd4066b108a8af503c1bc08a1d88f2b177ea","abstract_canon_sha256":"2bc9771b1647a0edae9196a0c80d74e96ef7d8e56b4c0089a84fb6ac23f1fc3b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:01:13.245836Z","signature_b64":"4gXxbKceneRfdep45j3u7yrR9+OuA/4Kr30j8Ko9s1TsplKSnJus9xre6EsFMobHqJ3gjhs6ukMdDEjO4ZTjCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c352e3d4ba97c498e1dda7041b1ff80e568000a2521d4ff5c66bf985c68b5b31","last_reissued_at":"2026-07-05T12:01:13.245290Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:01:13.245290Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Quantum Verifiable Rewards for Post-Training Qiskit Code Assistant","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"quant-ph","authors_text":"Adarsh Tiwari, David Kremer, Ismael Faro, Juan Cruz-Benito, Nicolas Dupuis, Youssef Mroueh","submitted_at":"2025-08-28T15:37:40Z","abstract_excerpt":"Qiskit is an open-source quantum computing framework that allows users to design, simulate, and run quantum circuits on real quantum hardware. We explore post-training techniques for LLMs to assist in writing Qiskit code. We introduce quantum verification as an effective method for ensuring code quality and executability on quantum hardware. To support this, we developed a synthetic data pipeline that generates quantum problem-unit test pairs and used it to create preference data for aligning LLMs with DPO. Additionally, we trained models using GRPO, leveraging quantum-verifiable rewards provi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.20907","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/2508.20907/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":"2508.20907","created_at":"2026-07-05T12:01:13.245378+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.20907v1","created_at":"2026-07-05T12:01:13.245378+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.20907","created_at":"2026-07-05T12:01:13.245378+00:00"},{"alias_kind":"pith_short_12","alias_value":"YNJOHVF2S7CJ","created_at":"2026-07-05T12:01:13.245378+00:00"},{"alias_kind":"pith_short_16","alias_value":"YNJOHVF2S7CJRYO5","created_at":"2026-07-05T12:01:13.245378+00:00"},{"alias_kind":"pith_short_8","alias_value":"YNJOHVF2","created_at":"2026-07-05T12:01:13.245378+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21974","citing_title":"Fine-Tuning Large Language Models for Quantum Reasoning","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18422","citing_title":"Gatekeepers and Hallucinations: A Layered Evaluation Framework for LLM-Driven Quantum Circuit Generation","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YNJOHVF2S7CJRYO5U4CBWH7YBZ","json":"https://pith.science/pith/YNJOHVF2S7CJRYO5U4CBWH7YBZ.json","graph_json":"https://pith.science/api/pith-number/YNJOHVF2S7CJRYO5U4CBWH7YBZ/graph.json","events_json":"https://pith.science/api/pith-number/YNJOHVF2S7CJRYO5U4CBWH7YBZ/events.json","paper":"https://pith.science/paper/YNJOHVF2"},"agent_actions":{"view_html":"https://pith.science/pith/YNJOHVF2S7CJRYO5U4CBWH7YBZ","download_json":"https://pith.science/pith/YNJOHVF2S7CJRYO5U4CBWH7YBZ.json","view_paper":"https://pith.science/paper/YNJOHVF2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.20907&json=true","fetch_graph":"https://pith.science/api/pith-number/YNJOHVF2S7CJRYO5U4CBWH7YBZ/graph.json","fetch_events":"https://pith.science/api/pith-number/YNJOHVF2S7CJRYO5U4CBWH7YBZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YNJOHVF2S7CJRYO5U4CBWH7YBZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YNJOHVF2S7CJRYO5U4CBWH7YBZ/action/storage_attestation","attest_author":"https://pith.science/pith/YNJOHVF2S7CJRYO5U4CBWH7YBZ/action/author_attestation","sign_citation":"https://pith.science/pith/YNJOHVF2S7CJRYO5U4CBWH7YBZ/action/citation_signature","submit_replication":"https://pith.science/pith/YNJOHVF2S7CJRYO5U4CBWH7YBZ/action/replication_record"}},"created_at":"2026-07-05T12:01:13.245378+00:00","updated_at":"2026-07-05T12:01:13.245378+00:00"}