{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:QOBGJW5GCJD6J2M6VLICK3P36L","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":"4cc8b1b258cd07073623b644b12c78e0896b6171ca4117c8432c530fb2020f12","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2026-06-11T03:48:18Z","title_canon_sha256":"49190d4ceefdcaf81e9968ad0ace315e25b689a05adbab436c193b242454ab7a"},"schema_version":"1.0","source":{"id":"2606.12864","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.12864","created_at":"2026-06-12T01:08:55Z"},{"alias_kind":"arxiv_version","alias_value":"2606.12864v1","created_at":"2026-06-12T01:08:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.12864","created_at":"2026-06-12T01:08:55Z"},{"alias_kind":"pith_short_12","alias_value":"QOBGJW5GCJD6","created_at":"2026-06-12T01:08:55Z"},{"alias_kind":"pith_short_16","alias_value":"QOBGJW5GCJD6J2M6","created_at":"2026-06-12T01:08:55Z"},{"alias_kind":"pith_short_8","alias_value":"QOBGJW5G","created_at":"2026-06-12T01:08:55Z"}],"graph_snapshots":[{"event_id":"sha256:9f6ff1083aff60fae884e03c8c1beea33da80b34a4865c2651f6dd6f7aa73b48","target":"graph","created_at":"2026-06-12T01:08:55Z","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/2606.12864/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Despite strong performance in competitive programming, the role of Large Language Models (LLMs) in supporting human learning in the same setting remains largely unexplored. In this work, we introduce UOJ-Bench, a benchmark designed to evaluate not only the problem-solving ability of LLMs, but also their ability to identify errors in human-written code -- a crucial educational activity traditionally supported by running test cases over online judge systems. UOJ-Bench consists of three distinct tasks: code generation, code hacking, and code repair, all constructed from real-world code submission","authors_text":"Alex Gu, Hangrui Zhou, Kaifeng Lyu, Tianle Cai, Tingqiang Xu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2026-06-11T03:48:18Z","title":"Beyond Problem Solving: UOJ-Bench for Evaluating Code Generation, Hacking, and Repair in Competitive Programming"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.12864","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:b4adc76ac6a1dc5138c9e4881d78827fdd0d459567948622e655a9bbe1145e24","target":"record","created_at":"2026-06-12T01:08:55Z","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":"4cc8b1b258cd07073623b644b12c78e0896b6171ca4117c8432c530fb2020f12","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2026-06-11T03:48:18Z","title_canon_sha256":"49190d4ceefdcaf81e9968ad0ace315e25b689a05adbab436c193b242454ab7a"},"schema_version":"1.0","source":{"id":"2606.12864","kind":"arxiv","version":1}},"canonical_sha256":"838264dba61247e4e99eaad0256dfbf2dd8e677dab2fa02401f3e36ab2d34a9c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"838264dba61247e4e99eaad0256dfbf2dd8e677dab2fa02401f3e36ab2d34a9c","first_computed_at":"2026-06-12T01:08:55.511084Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-12T01:08:55.511084Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rVaSf2wZcPMvXCK0dymNxRz/IW5YDy55SogfmPqaBZaNU7HEyfUx5jaCrcI2gebN1k3GZzWXjtZpOM1gVAyFDw==","signature_status":"signed_v1","signed_at":"2026-06-12T01:08:55.511943Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.12864","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b4adc76ac6a1dc5138c9e4881d78827fdd0d459567948622e655a9bbe1145e24","sha256:9f6ff1083aff60fae884e03c8c1beea33da80b34a4865c2651f6dd6f7aa73b48"],"state_sha256":"e7983f5c027fbe6a8f6cc6a17195e5fe31d77c1a2fffafae73896b477452930d"}