{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:UQXNQCHGLMGUIYKVIVHRXNA6BZ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"b2b8500709567920e343637dcd377c461ba804ee85db4aa9c4977c8826b5dcea","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-28T09:12:42Z","title_canon_sha256":"c5f65438c05efeb772d3c71c6afb76ea992e83108784085a8a39de05175a29c1"},"schema_version":"1.0","source":{"id":"2502.20868","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.20868","created_at":"2026-07-05T10:21:41Z"},{"alias_kind":"arxiv_version","alias_value":"2502.20868v1","created_at":"2026-07-05T10:21:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20868","created_at":"2026-07-05T10:21:41Z"},{"alias_kind":"pith_short_12","alias_value":"UQXNQCHGLMGU","created_at":"2026-07-05T10:21:41Z"},{"alias_kind":"pith_short_16","alias_value":"UQXNQCHGLMGUIYKV","created_at":"2026-07-05T10:21:41Z"},{"alias_kind":"pith_short_8","alias_value":"UQXNQCHG","created_at":"2026-07-05T10:21:41Z"}],"graph_snapshots":[{"event_id":"sha256:b31db8f9d17eb2b7cd38b17aac5e305d313ccc49a0d6042d7f77b783719bea64","target":"graph","created_at":"2026-07-05T10:21:41Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2502.20868/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"With reasoning language models such as OpenAI-o3 and DeepSeek-R1 emerging, large language models (LLMs) have entered a new phase of development. However, existing benchmarks for coding evaluation are gradually inadequate to assess the capability of advanced LLMs in code reasoning. To bridge the gap for high-level code reasoning assessment, we propose ProBench to benchmark LLMs in competitive programming, drawing inspiration from the International Collegiate Programming Contest. ProBench collects a comprehensive set of competitive programming problems from Codeforces, Luogu, and Nowcoder platfo","authors_text":"Deyi Xiong, Jianxiang Peng, Lei Yang, Ling Shi, Renren Jin, Yue Chen","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-28T09:12:42Z","title":"ProBench: Benchmarking Large Language Models in Competitive Programming"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20868","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:466f010c36aca90dece39040ca3d62f51e17ac35898e26b6c5340156ea98da24","target":"record","created_at":"2026-07-05T10:21:41Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"b2b8500709567920e343637dcd377c461ba804ee85db4aa9c4977c8826b5dcea","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-28T09:12:42Z","title_canon_sha256":"c5f65438c05efeb772d3c71c6afb76ea992e83108784085a8a39de05175a29c1"},"schema_version":"1.0","source":{"id":"2502.20868","kind":"arxiv","version":1}},"canonical_sha256":"a42ed808e65b0d446155454f1bb41e0e638356f3fc35bf9eca427be64f7db181","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a42ed808e65b0d446155454f1bb41e0e638356f3fc35bf9eca427be64f7db181","first_computed_at":"2026-07-05T10:21:41.416488Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:21:41.416488Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Y+uV5tzJuwJYqTub3L/OqPTkx9z7VzfDDtJM26hVzzm3dneDxmNMhIXBFEjvyXBXZAUDINUhYcpDpoqcHYvXCg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:21:41.417009Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.20868","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:466f010c36aca90dece39040ca3d62f51e17ac35898e26b6c5340156ea98da24","sha256:b31db8f9d17eb2b7cd38b17aac5e305d313ccc49a0d6042d7f77b783719bea64"],"state_sha256":"0513992bf1fc6451d212f9d6097673450cf407775bd7ef6eb4436c10eee59acf"}