{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SWUXT6IZLAP6NBXCM2FWVF2JG3","short_pith_number":"pith:SWUXT6IZ","schema_version":"1.0","canonical_sha256":"95a979f919581fe686e2668b6a974936fd5aaea21fa85ee134b505ac92db54b0","source":{"kind":"arxiv","id":"2501.12934","version":2},"attestation_state":"computed","paper":{"title":"Correctness Assessment of Code Generated by Large Language Models Using Internal Representations","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.SE","authors_text":"Hieu Dinh Vo, Son Nguyen, Thanh Trong Vu, Thu-Trang Nguyen, Tuan-Dung Bui","submitted_at":"2025-01-22T15:04:13Z","abstract_excerpt":"Ensuring the correctness of code generated by Large Language Models (LLMs) presents a significant challenge in AI-driven software development. Existing approaches predominantly rely on black-box (closed-box) approaches that evaluate correctness post-generation, failing to utilize the rich insights embedded in the LLMs' internal states during code generation. In this paper, we introduce OPENIA, a novel white-box (open-box) framework that leverages these internal representations to assess the correctness of LLM-generated code. OPENIA systematically analyzes the intermediate states of representat"},"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":"2501.12934","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.SE","submitted_at":"2025-01-22T15:04:13Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"f9febac673f4b1cb3a87e983704a84031c3aea56b9b208a1456ca3b859f9e73f","abstract_canon_sha256":"3545ea538c0f606b5c66e7c0ff0a0bc60faf40a079b98710ed7e6e45f96582ef"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:47:18.026694Z","signature_b64":"ivQ45/XPXBRnnDJQedrgML66XBkVDeBoBinuLUGgY/gEg63RfJWAQAZTgq75c3456DzR7ZWqoBjS+0c1oBcRDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"95a979f919581fe686e2668b6a974936fd5aaea21fa85ee134b505ac92db54b0","last_reissued_at":"2026-07-05T11:47:18.026089Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:47:18.026089Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Correctness Assessment of Code Generated by Large Language Models Using Internal Representations","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.SE","authors_text":"Hieu Dinh Vo, Son Nguyen, Thanh Trong Vu, Thu-Trang Nguyen, Tuan-Dung Bui","submitted_at":"2025-01-22T15:04:13Z","abstract_excerpt":"Ensuring the correctness of code generated by Large Language Models (LLMs) presents a significant challenge in AI-driven software development. Existing approaches predominantly rely on black-box (closed-box) approaches that evaluate correctness post-generation, failing to utilize the rich insights embedded in the LLMs' internal states during code generation. In this paper, we introduce OPENIA, a novel white-box (open-box) framework that leverages these internal representations to assess the correctness of LLM-generated code. OPENIA systematically analyzes the intermediate states of representat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.12934","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/2501.12934/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":"2501.12934","created_at":"2026-07-05T11:47:18.026157+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.12934v2","created_at":"2026-07-05T11:47:18.026157+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.12934","created_at":"2026-07-05T11:47:18.026157+00:00"},{"alias_kind":"pith_short_12","alias_value":"SWUXT6IZLAP6","created_at":"2026-07-05T11:47:18.026157+00:00"},{"alias_kind":"pith_short_16","alias_value":"SWUXT6IZLAP6NBXC","created_at":"2026-07-05T11:47:18.026157+00:00"},{"alias_kind":"pith_short_8","alias_value":"SWUXT6IZ","created_at":"2026-07-05T11:47:18.026157+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.06214","citing_title":"Towards the Readability of LLM-Generated Codes through Multitask Representation Engineering","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20211","citing_title":"Towards Secure Logging: Characterizing and Benchmarking Logging Code Security Issues with LLMs","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SWUXT6IZLAP6NBXCM2FWVF2JG3","json":"https://pith.science/pith/SWUXT6IZLAP6NBXCM2FWVF2JG3.json","graph_json":"https://pith.science/api/pith-number/SWUXT6IZLAP6NBXCM2FWVF2JG3/graph.json","events_json":"https://pith.science/api/pith-number/SWUXT6IZLAP6NBXCM2FWVF2JG3/events.json","paper":"https://pith.science/paper/SWUXT6IZ"},"agent_actions":{"view_html":"https://pith.science/pith/SWUXT6IZLAP6NBXCM2FWVF2JG3","download_json":"https://pith.science/pith/SWUXT6IZLAP6NBXCM2FWVF2JG3.json","view_paper":"https://pith.science/paper/SWUXT6IZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.12934&json=true","fetch_graph":"https://pith.science/api/pith-number/SWUXT6IZLAP6NBXCM2FWVF2JG3/graph.json","fetch_events":"https://pith.science/api/pith-number/SWUXT6IZLAP6NBXCM2FWVF2JG3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SWUXT6IZLAP6NBXCM2FWVF2JG3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SWUXT6IZLAP6NBXCM2FWVF2JG3/action/storage_attestation","attest_author":"https://pith.science/pith/SWUXT6IZLAP6NBXCM2FWVF2JG3/action/author_attestation","sign_citation":"https://pith.science/pith/SWUXT6IZLAP6NBXCM2FWVF2JG3/action/citation_signature","submit_replication":"https://pith.science/pith/SWUXT6IZLAP6NBXCM2FWVF2JG3/action/replication_record"}},"created_at":"2026-07-05T11:47:18.026157+00:00","updated_at":"2026-07-05T11:47:18.026157+00:00"}