{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:XEUSYR63NT3D4OJAIB6U5LBRQT","short_pith_number":"pith:XEUSYR63","canonical_record":{"source":{"id":"2506.04894","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T11:20:37Z","cross_cats_sorted":[],"title_canon_sha256":"51fc96a27edfe7d4288c2fb6c7db41a5c1a08091bb1ddd93ff2ebbd1006b58d2","abstract_canon_sha256":"b10e021bf7056bc44ac37ae834f82d1219985857aff4fffd82ce4f3268fb092e"},"schema_version":"1.0"},"canonical_sha256":"b9292c47db6cf63e3920407d4eac3184e7193b3bb8d36dd07f476cd3b8997e8e","source":{"kind":"arxiv","id":"2506.04894","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.04894","created_at":"2026-07-05T11:16:38Z"},{"alias_kind":"arxiv_version","alias_value":"2506.04894v1","created_at":"2026-07-05T11:16:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.04894","created_at":"2026-07-05T11:16:38Z"},{"alias_kind":"pith_short_12","alias_value":"XEUSYR63NT3D","created_at":"2026-07-05T11:16:38Z"},{"alias_kind":"pith_short_16","alias_value":"XEUSYR63NT3D4OJA","created_at":"2026-07-05T11:16:38Z"},{"alias_kind":"pith_short_8","alias_value":"XEUSYR63","created_at":"2026-07-05T11:16:38Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:XEUSYR63NT3D4OJAIB6U5LBRQT","target":"record","payload":{"canonical_record":{"source":{"id":"2506.04894","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T11:20:37Z","cross_cats_sorted":[],"title_canon_sha256":"51fc96a27edfe7d4288c2fb6c7db41a5c1a08091bb1ddd93ff2ebbd1006b58d2","abstract_canon_sha256":"b10e021bf7056bc44ac37ae834f82d1219985857aff4fffd82ce4f3268fb092e"},"schema_version":"1.0"},"canonical_sha256":"b9292c47db6cf63e3920407d4eac3184e7193b3bb8d36dd07f476cd3b8997e8e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:16:38.682393Z","signature_b64":"LJOtYB1cU4hoejJilVTrrq2rGgvjKYOGIrXJ1drlMH+5XCksPH9YbgTgwF/YGchf9wCJkqfHXNo8fJ1/6GpUBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b9292c47db6cf63e3920407d4eac3184e7193b3bb8d36dd07f476cd3b8997e8e","last_reissued_at":"2026-07-05T11:16:38.681846Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:16:38.681846Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.04894","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:16:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eg0DmMyW9C206y3C/cCgVVqylRigCNh/cDl0v4n2JxHmATajHlRTKxqsg0RN9LGI8wzMqd38lOEL82+2yS5mDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T03:40:10.759425Z"},"content_sha256":"9051075c072d0dbeaf9cbece307960d0227e1ba183a1cd25d33f9558a0b67091","schema_version":"1.0","event_id":"sha256:9051075c072d0dbeaf9cbece307960d0227e1ba183a1cd25d33f9558a0b67091"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:XEUSYR63NT3D4OJAIB6U5LBRQT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ICPC-Eval: Probing the Frontiers of LLM Reasoning with Competitive Programming Contests","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ji-Rong Wen, Shiyi Xu, Wayne Xin Zhao, Yingqian Min, Yiwen Hu, Zhipeng Chen","submitted_at":"2025-06-05T11:20:37Z","abstract_excerpt":"With the significant progress of large reasoning models in complex coding and reasoning tasks, existing benchmarks, like LiveCodeBench and CodeElo, are insufficient to evaluate the coding capabilities of large language models (LLMs) in real competition environments. Moreover, current evaluation metrics such as Pass@K fail to capture the reflective abilities of reasoning models. To address these challenges, we propose \\textbf{ICPC-Eval}, a top-level competitive coding benchmark designed to probing the frontiers of LLM reasoning. ICPC-Eval includes 118 carefully curated problems from 11 recent I"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.04894","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/2506.04894/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:16:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dRiy+zlRuGlWaNX988EgMbu+R30w1dJOnW5pxMoHlmpYc/MxStiq6yCMpcbuIBhTnSkdjpkYYM836thVhlLFDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T03:40:10.759917Z"},"content_sha256":"7d5aaeae673c19928b78c616d8a4b62f0057f447557795081fd8fb2c890ffac0","schema_version":"1.0","event_id":"sha256:7d5aaeae673c19928b78c616d8a4b62f0057f447557795081fd8fb2c890ffac0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XEUSYR63NT3D4OJAIB6U5LBRQT/bundle.json","state_url":"https://pith.science/pith/XEUSYR63NT3D4OJAIB6U5LBRQT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XEUSYR63NT3D4OJAIB6U5LBRQT/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-08T03:40:10Z","links":{"resolver":"https://pith.science/pith/XEUSYR63NT3D4OJAIB6U5LBRQT","bundle":"https://pith.science/pith/XEUSYR63NT3D4OJAIB6U5LBRQT/bundle.json","state":"https://pith.science/pith/XEUSYR63NT3D4OJAIB6U5LBRQT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XEUSYR63NT3D4OJAIB6U5LBRQT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:XEUSYR63NT3D4OJAIB6U5LBRQT","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":"b10e021bf7056bc44ac37ae834f82d1219985857aff4fffd82ce4f3268fb092e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T11:20:37Z","title_canon_sha256":"51fc96a27edfe7d4288c2fb6c7db41a5c1a08091bb1ddd93ff2ebbd1006b58d2"},"schema_version":"1.0","source":{"id":"2506.04894","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.04894","created_at":"2026-07-05T11:16:38Z"},{"alias_kind":"arxiv_version","alias_value":"2506.04894v1","created_at":"2026-07-05T11:16:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.04894","created_at":"2026-07-05T11:16:38Z"},{"alias_kind":"pith_short_12","alias_value":"XEUSYR63NT3D","created_at":"2026-07-05T11:16:38Z"},{"alias_kind":"pith_short_16","alias_value":"XEUSYR63NT3D4OJA","created_at":"2026-07-05T11:16:38Z"},{"alias_kind":"pith_short_8","alias_value":"XEUSYR63","created_at":"2026-07-05T11:16:38Z"}],"graph_snapshots":[{"event_id":"sha256:7d5aaeae673c19928b78c616d8a4b62f0057f447557795081fd8fb2c890ffac0","target":"graph","created_at":"2026-07-05T11:16:38Z","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/2506.04894/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"With the significant progress of large reasoning models in complex coding and reasoning tasks, existing benchmarks, like LiveCodeBench and CodeElo, are insufficient to evaluate the coding capabilities of large language models (LLMs) in real competition environments. Moreover, current evaluation metrics such as Pass@K fail to capture the reflective abilities of reasoning models. To address these challenges, we propose \\textbf{ICPC-Eval}, a top-level competitive coding benchmark designed to probing the frontiers of LLM reasoning. ICPC-Eval includes 118 carefully curated problems from 11 recent I","authors_text":"Ji-Rong Wen, Shiyi Xu, Wayne Xin Zhao, Yingqian Min, Yiwen Hu, Zhipeng Chen","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T11:20:37Z","title":"ICPC-Eval: Probing the Frontiers of LLM Reasoning with Competitive Programming Contests"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.04894","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":"sha256:9051075c072d0dbeaf9cbece307960d0227e1ba183a1cd25d33f9558a0b67091","target":"record","created_at":"2026-07-05T11:16:38Z","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":"b10e021bf7056bc44ac37ae834f82d1219985857aff4fffd82ce4f3268fb092e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T11:20:37Z","title_canon_sha256":"51fc96a27edfe7d4288c2fb6c7db41a5c1a08091bb1ddd93ff2ebbd1006b58d2"},"schema_version":"1.0","source":{"id":"2506.04894","kind":"arxiv","version":1}},"canonical_sha256":"b9292c47db6cf63e3920407d4eac3184e7193b3bb8d36dd07f476cd3b8997e8e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b9292c47db6cf63e3920407d4eac3184e7193b3bb8d36dd07f476cd3b8997e8e","first_computed_at":"2026-07-05T11:16:38.681846Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:16:38.681846Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"LJOtYB1cU4hoejJilVTrrq2rGgvjKYOGIrXJ1drlMH+5XCksPH9YbgTgwF/YGchf9wCJkqfHXNo8fJ1/6GpUBg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:16:38.682393Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.04894","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9051075c072d0dbeaf9cbece307960d0227e1ba183a1cd25d33f9558a0b67091","sha256:7d5aaeae673c19928b78c616d8a4b62f0057f447557795081fd8fb2c890ffac0"],"state_sha256":"c252314560d0012f5a396fc7b007f36472d3625caa344d54401a092340c39342"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3mC+rH0JXeYpg76Ybwnnx/vLr9AcoD6fqYB9tdjqxOpZ5XexfUqB6x12ECU2XEiqg05NGmJt78AGfQ3sAs/LBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T03:40:10.763843Z","bundle_sha256":"6e870bfd33d6cbbcf03c8f745d235a61e8761b6c843d3228f1be881d1fea60ac"}}