{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:Z6NPIFOL5NBYLSZOJMNGJRERFZ","short_pith_number":"pith:Z6NPIFOL","schema_version":"1.0","canonical_sha256":"cf9af415cbeb4385cb2e4b1a64c4912e480836aa55efced33aee014b992ea692","source":{"kind":"arxiv","id":"2502.18487","version":1},"attestation_state":"computed","paper":{"title":"AuPair: Golden Example Pairs for Code Repair","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.SE","authors_text":"Aditi Mavalankar, Hassan Mansoor, Masha Samsikova, Tom Schaul, Zita Marinho","submitted_at":"2025-02-12T11:07:04Z","abstract_excerpt":"Scaling up inference-time compute has proven to be a valuable strategy in improving the performance of Large Language Models (LLMs) without fine-tuning. An important task that can benefit from additional inference-time compute is self-repair; given an initial flawed response, or guess, the LLM corrects its own mistake and produces an improved response, or fix. We leverage the in-context learning ability of LLMs to perform self-repair in the coding domain. The key contribution of our paper is an approach that synthesises and selects an ordered set of golden example pairs, or AuPairs, of these i"},"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":"2502.18487","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2025-02-12T11:07:04Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"ad3b5e19ecdf01c2f93e58b2d29bb0139e144ff61df3c9f2de6238b4865f59a4","abstract_canon_sha256":"7568408a63db2a3015be25614cc41e25c77caa72275c0307c9d0880cdf546b70"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:19:58.703226Z","signature_b64":"bwyWijHAJmMhTi4UsZXda08XysE9/H9sHxTOr/jpRKhuVo03D/zEFSG97jHPeZuFrfTsnhp5aVVtSK+XQpFiCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cf9af415cbeb4385cb2e4b1a64c4912e480836aa55efced33aee014b992ea692","last_reissued_at":"2026-07-05T10:19:58.702726Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:19:58.702726Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AuPair: Golden Example Pairs for Code Repair","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.SE","authors_text":"Aditi Mavalankar, Hassan Mansoor, Masha Samsikova, Tom Schaul, Zita Marinho","submitted_at":"2025-02-12T11:07:04Z","abstract_excerpt":"Scaling up inference-time compute has proven to be a valuable strategy in improving the performance of Large Language Models (LLMs) without fine-tuning. An important task that can benefit from additional inference-time compute is self-repair; given an initial flawed response, or guess, the LLM corrects its own mistake and produces an improved response, or fix. We leverage the in-context learning ability of LLMs to perform self-repair in the coding domain. The key contribution of our paper is an approach that synthesises and selects an ordered set of golden example pairs, or AuPairs, of these i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.18487","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/2502.18487/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":"2502.18487","created_at":"2026-07-05T10:19:58.702789+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.18487v1","created_at":"2026-07-05T10:19:58.702789+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.18487","created_at":"2026-07-05T10:19:58.702789+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z6NPIFOL5NBY","created_at":"2026-07-05T10:19:58.702789+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z6NPIFOL5NBYLSZO","created_at":"2026-07-05T10:19:58.702789+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z6NPIFOL","created_at":"2026-07-05T10:19:58.702789+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.02330","citing_title":"ReCode: Improving LLM-based Code Repair with Fine-Grained Retrieval-Augmented Generation","ref_index":29,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Z6NPIFOL5NBYLSZOJMNGJRERFZ","json":"https://pith.science/pith/Z6NPIFOL5NBYLSZOJMNGJRERFZ.json","graph_json":"https://pith.science/api/pith-number/Z6NPIFOL5NBYLSZOJMNGJRERFZ/graph.json","events_json":"https://pith.science/api/pith-number/Z6NPIFOL5NBYLSZOJMNGJRERFZ/events.json","paper":"https://pith.science/paper/Z6NPIFOL"},"agent_actions":{"view_html":"https://pith.science/pith/Z6NPIFOL5NBYLSZOJMNGJRERFZ","download_json":"https://pith.science/pith/Z6NPIFOL5NBYLSZOJMNGJRERFZ.json","view_paper":"https://pith.science/paper/Z6NPIFOL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.18487&json=true","fetch_graph":"https://pith.science/api/pith-number/Z6NPIFOL5NBYLSZOJMNGJRERFZ/graph.json","fetch_events":"https://pith.science/api/pith-number/Z6NPIFOL5NBYLSZOJMNGJRERFZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z6NPIFOL5NBYLSZOJMNGJRERFZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z6NPIFOL5NBYLSZOJMNGJRERFZ/action/storage_attestation","attest_author":"https://pith.science/pith/Z6NPIFOL5NBYLSZOJMNGJRERFZ/action/author_attestation","sign_citation":"https://pith.science/pith/Z6NPIFOL5NBYLSZOJMNGJRERFZ/action/citation_signature","submit_replication":"https://pith.science/pith/Z6NPIFOL5NBYLSZOJMNGJRERFZ/action/replication_record"}},"created_at":"2026-07-05T10:19:58.702789+00:00","updated_at":"2026-07-05T10:19:58.702789+00:00"}