{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:IQB3M75BQWPIOS4G4GZRQFERP2","short_pith_number":"pith:IQB3M75B","schema_version":"1.0","canonical_sha256":"4403b67fa1859e874b86e1b31814917eace9a6df79a1f16a00deae079e3a028d","source":{"kind":"arxiv","id":"2403.10059","version":2},"attestation_state":"computed","paper":{"title":"Repoformer: Selective Retrieval for Repository-Level Code Completion","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SE","authors_text":"Dejiao Zhang, Di Wu, Murali Krishna Ramanathan, Wasi Uddin Ahmad, Xiaofei Ma","submitted_at":"2024-03-15T06:59:43Z","abstract_excerpt":"Recent advances in retrieval-augmented generation (RAG) have initiated a new era in repository-level code completion. However, the invariable use of retrieval in existing methods exposes issues in both efficiency and robustness, with a large proportion of the retrieved contexts proving unhelpful or harmful to code language models (code LMs). In this paper, we propose a selective RAG framework to avoid retrieval when unnecessary. To power this framework, we design a self-supervised learning approach to enable a code LM to accurately self-evaluate whether retrieval can improve its output quality"},"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":"2403.10059","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2024-03-15T06:59:43Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"0008fb3335fea2d621c920f53495145e3781c893ad1242a195983341d6d2a177","abstract_canon_sha256":"c82b6e98620067ae308b0794bb1f3c0edafdd986931e7e0875fa380db5dbf495"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:27:06.397055Z","signature_b64":"suruvF2Fu6r7doBCA/RWTGE0Su700AA0UlBUez03DyMiL1UJYQfNu8ubnDCnLvvg3CNhKBBmHyfyWbT29nvVBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4403b67fa1859e874b86e1b31814917eace9a6df79a1f16a00deae079e3a028d","last_reissued_at":"2026-07-05T08:27:06.396622Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:27:06.396622Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Repoformer: Selective Retrieval for Repository-Level Code Completion","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SE","authors_text":"Dejiao Zhang, Di Wu, Murali Krishna Ramanathan, Wasi Uddin Ahmad, Xiaofei Ma","submitted_at":"2024-03-15T06:59:43Z","abstract_excerpt":"Recent advances in retrieval-augmented generation (RAG) have initiated a new era in repository-level code completion. However, the invariable use of retrieval in existing methods exposes issues in both efficiency and robustness, with a large proportion of the retrieved contexts proving unhelpful or harmful to code language models (code LMs). In this paper, we propose a selective RAG framework to avoid retrieval when unnecessary. To power this framework, we design a self-supervised learning approach to enable a code LM to accurately self-evaluate whether retrieval can improve its output quality"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.10059","kind":"arxiv","version":2},"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/2403.10059/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":"2403.10059","created_at":"2026-07-05T08:27:06.396678+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.10059v2","created_at":"2026-07-05T08:27:06.396678+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.10059","created_at":"2026-07-05T08:27:06.396678+00:00"},{"alias_kind":"pith_short_12","alias_value":"IQB3M75BQWPI","created_at":"2026-07-05T08:27:06.396678+00:00"},{"alias_kind":"pith_short_16","alias_value":"IQB3M75BQWPIOS4G","created_at":"2026-07-05T08:27:06.396678+00:00"},{"alias_kind":"pith_short_8","alias_value":"IQB3M75B","created_at":"2026-07-05T08:27:06.396678+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.06492","citing_title":"Code2LoRA: Hypernetwork-Generated Adapters for Code Language Models under Software Evolution","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05720","citing_title":"Microskill Architecture: A Modular Skill-Driven Framework for AI-Native Code Generation","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2502.20295","citing_title":"Judge a Book by its Cover: Investigating Multi-Modal LLMs for Multi-Page Handwritten Document Transcription","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14478","citing_title":"When Retrieval Hurts Code Completion: A Diagnostic Study of Stale Repository Context","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02079","citing_title":"Automated Functional Testing for Malleable Mobile Application Driven from User Intent","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2406.00515","citing_title":"A Survey on Large Language Models for Code Generation","ref_index":289,"is_internal_anchor":false},{"citing_arxiv_id":"2409.12186","citing_title":"Qwen2.5-Coder Technical Report","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04763","citing_title":"How Does Chunking Affect Retrieval-Augmented Code Completion? A Controlled Empirical Study","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IQB3M75BQWPIOS4G4GZRQFERP2","json":"https://pith.science/pith/IQB3M75BQWPIOS4G4GZRQFERP2.json","graph_json":"https://pith.science/api/pith-number/IQB3M75BQWPIOS4G4GZRQFERP2/graph.json","events_json":"https://pith.science/api/pith-number/IQB3M75BQWPIOS4G4GZRQFERP2/events.json","paper":"https://pith.science/paper/IQB3M75B"},"agent_actions":{"view_html":"https://pith.science/pith/IQB3M75BQWPIOS4G4GZRQFERP2","download_json":"https://pith.science/pith/IQB3M75BQWPIOS4G4GZRQFERP2.json","view_paper":"https://pith.science/paper/IQB3M75B","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.10059&json=true","fetch_graph":"https://pith.science/api/pith-number/IQB3M75BQWPIOS4G4GZRQFERP2/graph.json","fetch_events":"https://pith.science/api/pith-number/IQB3M75BQWPIOS4G4GZRQFERP2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IQB3M75BQWPIOS4G4GZRQFERP2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IQB3M75BQWPIOS4G4GZRQFERP2/action/storage_attestation","attest_author":"https://pith.science/pith/IQB3M75BQWPIOS4G4GZRQFERP2/action/author_attestation","sign_citation":"https://pith.science/pith/IQB3M75BQWPIOS4G4GZRQFERP2/action/citation_signature","submit_replication":"https://pith.science/pith/IQB3M75BQWPIOS4G4GZRQFERP2/action/replication_record"}},"created_at":"2026-07-05T08:27:06.396678+00:00","updated_at":"2026-07-05T08:27:06.396678+00:00"}