{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:456F6KSNSL5OPDRJNF2JUICSJG","short_pith_number":"pith:456F6KSN","schema_version":"1.0","canonical_sha256":"e77c5f2a4d92fae78e2969749a205249b6e6c8508c3068365b8faf8ba798a0c1","source":{"kind":"arxiv","id":"2505.24098","version":1},"attestation_state":"computed","paper":{"title":"HardTests: Synthesizing High-Quality Test Cases for LLM Coding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aidan Zhang, Dejia Xu, Ivan Bercovich, Jiabao Ji, Junting Zhou, Kexun Zhang, Lei Li, Yee Man Choi, Zhongmou He","submitted_at":"2025-05-30T01:00:34Z","abstract_excerpt":"Verifiers play a crucial role in large language model (LLM) reasoning, needed by post-training techniques such as reinforcement learning. However, reliable verifiers are hard to get for difficult coding problems, because a well-disguised wrong solution may only be detected by carefully human-written edge cases that are difficult to synthesize. To address this issue, we propose HARDTESTGEN, a pipeline for high-quality test synthesis using LLMs. With this pipeline, we curate a comprehensive competitive programming dataset HARDTESTS with 47k problems and synthetic high-quality tests. Compared wit"},"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":"2505.24098","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-30T01:00:34Z","cross_cats_sorted":[],"title_canon_sha256":"84bbd61baf4c801fefadf5afb69be9ae3cd68892e8935ed3cd580a386dc81cbb","abstract_canon_sha256":"1091f49d455e8fa9266f5d0db9892c0f60266819d6e1e9c60b09be25d9fc6116"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:41.854763Z","signature_b64":"EX00apbVWzaaxzut8ZGY9cIJUv3ldRiDznXlXgjBP770Sw1kjArNDYtr58NlE6Ir+Wiw1GY8w34SbSIWyGUNAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e77c5f2a4d92fae78e2969749a205249b6e6c8508c3068365b8faf8ba798a0c1","last_reissued_at":"2026-07-05T11:12:41.854207Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:41.854207Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HardTests: Synthesizing High-Quality Test Cases for LLM Coding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aidan Zhang, Dejia Xu, Ivan Bercovich, Jiabao Ji, Junting Zhou, Kexun Zhang, Lei Li, Yee Man Choi, Zhongmou He","submitted_at":"2025-05-30T01:00:34Z","abstract_excerpt":"Verifiers play a crucial role in large language model (LLM) reasoning, needed by post-training techniques such as reinforcement learning. However, reliable verifiers are hard to get for difficult coding problems, because a well-disguised wrong solution may only be detected by carefully human-written edge cases that are difficult to synthesize. To address this issue, we propose HARDTESTGEN, a pipeline for high-quality test synthesis using LLMs. With this pipeline, we curate a comprehensive competitive programming dataset HARDTESTS with 47k problems and synthetic high-quality tests. Compared wit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24098","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/2505.24098/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":"2505.24098","created_at":"2026-07-05T11:12:41.854275+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.24098v1","created_at":"2026-07-05T11:12:41.854275+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24098","created_at":"2026-07-05T11:12:41.854275+00:00"},{"alias_kind":"pith_short_12","alias_value":"456F6KSNSL5O","created_at":"2026-07-05T11:12:41.854275+00:00"},{"alias_kind":"pith_short_16","alias_value":"456F6KSNSL5OPDRJ","created_at":"2026-07-05T11:12:41.854275+00:00"},{"alias_kind":"pith_short_8","alias_value":"456F6KSN","created_at":"2026-07-05T11:12:41.854275+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25450","citing_title":"The Generalization Spectrum: A Chromatographic Approach to Evaluating Learning Algorithms","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2606.25450","citing_title":"The Generalization Spectrum: A Chromatographic Approach to Evaluating Learning Algorithms","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24693","citing_title":"CP-Agent: A Calibrated Risk-Controlled Agent for Feedback-Driven Competitive Programming","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14445","citing_title":"FrontierSmith: Synthesizing Open-Ended Coding Problems at Scale","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08553","citing_title":"VeriContest: A Competitive-Programming Benchmark for Verifiable Code Generation","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21138","citing_title":"Navigating the Clutter: Waypoint-Based Bi-Level Planning for Multi-Robot Systems","ref_index":118,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/456F6KSNSL5OPDRJNF2JUICSJG","json":"https://pith.science/pith/456F6KSNSL5OPDRJNF2JUICSJG.json","graph_json":"https://pith.science/api/pith-number/456F6KSNSL5OPDRJNF2JUICSJG/graph.json","events_json":"https://pith.science/api/pith-number/456F6KSNSL5OPDRJNF2JUICSJG/events.json","paper":"https://pith.science/paper/456F6KSN"},"agent_actions":{"view_html":"https://pith.science/pith/456F6KSNSL5OPDRJNF2JUICSJG","download_json":"https://pith.science/pith/456F6KSNSL5OPDRJNF2JUICSJG.json","view_paper":"https://pith.science/paper/456F6KSN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.24098&json=true","fetch_graph":"https://pith.science/api/pith-number/456F6KSNSL5OPDRJNF2JUICSJG/graph.json","fetch_events":"https://pith.science/api/pith-number/456F6KSNSL5OPDRJNF2JUICSJG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/456F6KSNSL5OPDRJNF2JUICSJG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/456F6KSNSL5OPDRJNF2JUICSJG/action/storage_attestation","attest_author":"https://pith.science/pith/456F6KSNSL5OPDRJNF2JUICSJG/action/author_attestation","sign_citation":"https://pith.science/pith/456F6KSNSL5OPDRJNF2JUICSJG/action/citation_signature","submit_replication":"https://pith.science/pith/456F6KSNSL5OPDRJNF2JUICSJG/action/replication_record"}},"created_at":"2026-07-05T11:12:41.854275+00:00","updated_at":"2026-07-05T11:12:41.854275+00:00"}