{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XX2V66XTL6UQWPJPQSB2OC2PRH","short_pith_number":"pith:XX2V66XT","schema_version":"1.0","canonical_sha256":"bdf55f7af35fa90b3d2f8483a70b4f89f18387f0b2a631f5b5cdc53deeec7835","source":{"kind":"arxiv","id":"2508.07434","version":1},"attestation_state":"computed","paper":{"title":"Let's Revise Step-by-Step: A Unified Local Search Framework for Code Generation with LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bo An, Jianguo Huang, Steven Hoi, Yanchen Deng, Zhiyi Lyu","submitted_at":"2025-08-10T17:11:56Z","abstract_excerpt":"Large Language Models (LLMs) with inference-time scaling techniques show promise for code generation, yet face notable efficiency and scalability challenges. Construction-based tree-search methods suffer from rapid growth in tree size, high token consumption, and lack of anytime property. In contrast, improvement-based methods offer better performance but often struggle with uninformative reward signals and inefficient search strategies. In this work, we propose \\textbf{ReLoc}, a unified local search framework which effectively performs step-by-step code revision. Specifically, ReLoc explores "},"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":"2508.07434","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-10T17:11:56Z","cross_cats_sorted":[],"title_canon_sha256":"c721d9000b4b95ce27db3f1880e1e52a5e333c7146b7ad678e02d3831f170584","abstract_canon_sha256":"33c4726e6f69bd4f2c1c4c99cb2c8f171972bc78bd7438d604823b98459ba3a4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:51:46.221585Z","signature_b64":"3oKwXDAjdSu2YrdUwuP0T2D0vQag8EgYG0KVUixW3TSwM4N/JQtTVRm/62PAI1ZNNbjOpbYNmChQn1oqixTyCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bdf55f7af35fa90b3d2f8483a70b4f89f18387f0b2a631f5b5cdc53deeec7835","last_reissued_at":"2026-07-05T11:51:46.221068Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:51:46.221068Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Let's Revise Step-by-Step: A Unified Local Search Framework for Code Generation with LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bo An, Jianguo Huang, Steven Hoi, Yanchen Deng, Zhiyi Lyu","submitted_at":"2025-08-10T17:11:56Z","abstract_excerpt":"Large Language Models (LLMs) with inference-time scaling techniques show promise for code generation, yet face notable efficiency and scalability challenges. Construction-based tree-search methods suffer from rapid growth in tree size, high token consumption, and lack of anytime property. In contrast, improvement-based methods offer better performance but often struggle with uninformative reward signals and inefficient search strategies. In this work, we propose \\textbf{ReLoc}, a unified local search framework which effectively performs step-by-step code revision. Specifically, ReLoc explores "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.07434","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/2508.07434/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":"2508.07434","created_at":"2026-07-05T11:51:46.221128+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.07434v1","created_at":"2026-07-05T11:51:46.221128+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.07434","created_at":"2026-07-05T11:51:46.221128+00:00"},{"alias_kind":"pith_short_12","alias_value":"XX2V66XTL6UQ","created_at":"2026-07-05T11:51:46.221128+00:00"},{"alias_kind":"pith_short_16","alias_value":"XX2V66XTL6UQWPJP","created_at":"2026-07-05T11:51:46.221128+00:00"},{"alias_kind":"pith_short_8","alias_value":"XX2V66XT","created_at":"2026-07-05T11:51:46.221128+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.18747","citing_title":"Code as Agent Harness","ref_index":164,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10449","citing_title":"AdverMCTS: Combating Pseudo-Correctness in Code Generation via Adversarial Monte Carlo Tree Search","ref_index":36,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XX2V66XTL6UQWPJPQSB2OC2PRH","json":"https://pith.science/pith/XX2V66XTL6UQWPJPQSB2OC2PRH.json","graph_json":"https://pith.science/api/pith-number/XX2V66XTL6UQWPJPQSB2OC2PRH/graph.json","events_json":"https://pith.science/api/pith-number/XX2V66XTL6UQWPJPQSB2OC2PRH/events.json","paper":"https://pith.science/paper/XX2V66XT"},"agent_actions":{"view_html":"https://pith.science/pith/XX2V66XTL6UQWPJPQSB2OC2PRH","download_json":"https://pith.science/pith/XX2V66XTL6UQWPJPQSB2OC2PRH.json","view_paper":"https://pith.science/paper/XX2V66XT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.07434&json=true","fetch_graph":"https://pith.science/api/pith-number/XX2V66XTL6UQWPJPQSB2OC2PRH/graph.json","fetch_events":"https://pith.science/api/pith-number/XX2V66XTL6UQWPJPQSB2OC2PRH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XX2V66XTL6UQWPJPQSB2OC2PRH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XX2V66XTL6UQWPJPQSB2OC2PRH/action/storage_attestation","attest_author":"https://pith.science/pith/XX2V66XTL6UQWPJPQSB2OC2PRH/action/author_attestation","sign_citation":"https://pith.science/pith/XX2V66XTL6UQWPJPQSB2OC2PRH/action/citation_signature","submit_replication":"https://pith.science/pith/XX2V66XTL6UQWPJPQSB2OC2PRH/action/replication_record"}},"created_at":"2026-07-05T11:51:46.221128+00:00","updated_at":"2026-07-05T11:51:46.221128+00:00"}