{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:643GD2UBUKDYLXOXV5IU6VRUSU","short_pith_number":"pith:643GD2UB","canonical_record":{"source":{"id":"2501.01257","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-02T13:49:00Z","cross_cats_sorted":[],"title_canon_sha256":"e0b31860b29d5230af3d5e8ff59497b5465951c7976da87790dfe3ab43cf74e2","abstract_canon_sha256":"c0a20f9968dfd32aadff58b71d0921165d6e309dfda3cc37bff9ee28efb8408b"},"schema_version":"1.0"},"canonical_sha256":"f73661ea81a28785ddd7af514f5634953e893826ef2c8bccecbe3b798b312a7a","source":{"kind":"arxiv","id":"2501.01257","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.01257","created_at":"2026-07-05T09:56:39Z"},{"alias_kind":"arxiv_version","alias_value":"2501.01257v2","created_at":"2026-07-05T09:56:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.01257","created_at":"2026-07-05T09:56:39Z"},{"alias_kind":"pith_short_12","alias_value":"643GD2UBUKDY","created_at":"2026-07-05T09:56:39Z"},{"alias_kind":"pith_short_16","alias_value":"643GD2UBUKDYLXOX","created_at":"2026-07-05T09:56:39Z"},{"alias_kind":"pith_short_8","alias_value":"643GD2UB","created_at":"2026-07-05T09:56:39Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:643GD2UBUKDYLXOXV5IU6VRUSU","target":"record","payload":{"canonical_record":{"source":{"id":"2501.01257","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-02T13:49:00Z","cross_cats_sorted":[],"title_canon_sha256":"e0b31860b29d5230af3d5e8ff59497b5465951c7976da87790dfe3ab43cf74e2","abstract_canon_sha256":"c0a20f9968dfd32aadff58b71d0921165d6e309dfda3cc37bff9ee28efb8408b"},"schema_version":"1.0"},"canonical_sha256":"f73661ea81a28785ddd7af514f5634953e893826ef2c8bccecbe3b798b312a7a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:56:39.221174Z","signature_b64":"ZzLRsjQEXliCPLoPcJERLF5oez3i/SWmSZPpauFRftI9igy/+k9n4LBWjis1bcU6NT3vrg/xP4FBtHJnoXXRBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f73661ea81a28785ddd7af514f5634953e893826ef2c8bccecbe3b798b312a7a","last_reissued_at":"2026-07-05T09:56:39.220669Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:56:39.220669Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.01257","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:56:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BLACNbxE67ipPjfdJBqJC4tn4H/L3R5FAKQo6W5j6ejTEej8HpzobNxu6xrzvOa078EQo6gcjON8wcoLYXRWAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T05:01:18.143742Z"},"content_sha256":"b84dccddce65b8672774a99560bac07e77337d10416b0a4462b968a7fc099966","schema_version":"1.0","event_id":"sha256:b84dccddce65b8672774a99560bac07e77337d10416b0a4462b968a7fc099966"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:643GD2UBUKDYLXOXV5IU6VRUSU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CodeElo: Benchmarking Competition-level Code Generation of LLMs with Human-comparable Elo Ratings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"An Yang, Binyuan Hui, Bofei Gao, Bowen Yu, Bo Zheng, Dayiheng Liu, Jian Yang, Jiaxi Yang, Junyang Lin, Shanghaoran Quan, Xuancheng Ren, Yang Fan, Yibo Miao, Yichang Zhang, Yunlong Feng, Zekun Wang, Zeyu Cui","submitted_at":"2025-01-02T13:49:00Z","abstract_excerpt":"With the increasing code reasoning capabilities of existing large language models (LLMs) and breakthroughs in reasoning models like OpenAI o1 and o3, there is a growing need to develop more challenging and comprehensive benchmarks that effectively test their sophisticated competition-level coding abilities. Existing benchmarks, like LiveCodeBench and USACO, fall short due to the unavailability of private test cases, lack of support for special judges, and misaligned execution environments. To bridge this gap, we introduce CodeElo, a standardized competition-level code generation benchmark that"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.01257","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/2501.01257/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":"2026-07-05T09:56:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LDmnlivg4amY8hEytgQdZAJ788PAMuqIr824dHUrAki687SW+50Zjo56PKbDSI7mDvYIxUj5job6LlolCxBdDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T05:01:18.144611Z"},"content_sha256":"1c5651cd8afdd98fca8e66e63dbb6aedcc72002925ac373fdb989932ec4745a0","schema_version":"1.0","event_id":"sha256:1c5651cd8afdd98fca8e66e63dbb6aedcc72002925ac373fdb989932ec4745a0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/643GD2UBUKDYLXOXV5IU6VRUSU/bundle.json","state_url":"https://pith.science/pith/643GD2UBUKDYLXOXV5IU6VRUSU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/643GD2UBUKDYLXOXV5IU6VRUSU/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":"2026-08-01T05:01:18Z","links":{"resolver":"https://pith.science/pith/643GD2UBUKDYLXOXV5IU6VRUSU","bundle":"https://pith.science/pith/643GD2UBUKDYLXOXV5IU6VRUSU/bundle.json","state":"https://pith.science/pith/643GD2UBUKDYLXOXV5IU6VRUSU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/643GD2UBUKDYLXOXV5IU6VRUSU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:643GD2UBUKDYLXOXV5IU6VRUSU","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":"c0a20f9968dfd32aadff58b71d0921165d6e309dfda3cc37bff9ee28efb8408b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-02T13:49:00Z","title_canon_sha256":"e0b31860b29d5230af3d5e8ff59497b5465951c7976da87790dfe3ab43cf74e2"},"schema_version":"1.0","source":{"id":"2501.01257","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.01257","created_at":"2026-07-05T09:56:39Z"},{"alias_kind":"arxiv_version","alias_value":"2501.01257v2","created_at":"2026-07-05T09:56:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.01257","created_at":"2026-07-05T09:56:39Z"},{"alias_kind":"pith_short_12","alias_value":"643GD2UBUKDY","created_at":"2026-07-05T09:56:39Z"},{"alias_kind":"pith_short_16","alias_value":"643GD2UBUKDYLXOX","created_at":"2026-07-05T09:56:39Z"},{"alias_kind":"pith_short_8","alias_value":"643GD2UB","created_at":"2026-07-05T09:56:39Z"}],"graph_snapshots":[{"event_id":"sha256:1c5651cd8afdd98fca8e66e63dbb6aedcc72002925ac373fdb989932ec4745a0","target":"graph","created_at":"2026-07-05T09:56:39Z","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/2501.01257/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"With the increasing code reasoning capabilities of existing large language models (LLMs) and breakthroughs in reasoning models like OpenAI o1 and o3, there is a growing need to develop more challenging and comprehensive benchmarks that effectively test their sophisticated competition-level coding abilities. Existing benchmarks, like LiveCodeBench and USACO, fall short due to the unavailability of private test cases, lack of support for special judges, and misaligned execution environments. To bridge this gap, we introduce CodeElo, a standardized competition-level code generation benchmark that","authors_text":"An Yang, Binyuan Hui, Bofei Gao, Bowen Yu, Bo Zheng, Dayiheng Liu, Jian Yang, Jiaxi Yang, Junyang Lin, Shanghaoran Quan, Xuancheng Ren, Yang Fan, Yibo Miao, Yichang Zhang, Yunlong Feng, Zekun Wang, Zeyu Cui","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-02T13:49:00Z","title":"CodeElo: Benchmarking Competition-level Code Generation of LLMs with Human-comparable Elo Ratings"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.01257","kind":"arxiv","version":2},"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:b84dccddce65b8672774a99560bac07e77337d10416b0a4462b968a7fc099966","target":"record","created_at":"2026-07-05T09:56:39Z","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":"c0a20f9968dfd32aadff58b71d0921165d6e309dfda3cc37bff9ee28efb8408b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-02T13:49:00Z","title_canon_sha256":"e0b31860b29d5230af3d5e8ff59497b5465951c7976da87790dfe3ab43cf74e2"},"schema_version":"1.0","source":{"id":"2501.01257","kind":"arxiv","version":2}},"canonical_sha256":"f73661ea81a28785ddd7af514f5634953e893826ef2c8bccecbe3b798b312a7a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f73661ea81a28785ddd7af514f5634953e893826ef2c8bccecbe3b798b312a7a","first_computed_at":"2026-07-05T09:56:39.220669Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:56:39.220669Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ZzLRsjQEXliCPLoPcJERLF5oez3i/SWmSZPpauFRftI9igy/+k9n4LBWjis1bcU6NT3vrg/xP4FBtHJnoXXRBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:56:39.221174Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.01257","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b84dccddce65b8672774a99560bac07e77337d10416b0a4462b968a7fc099966","sha256:1c5651cd8afdd98fca8e66e63dbb6aedcc72002925ac373fdb989932ec4745a0"],"state_sha256":"7af101bff1ca5cfef09bfe30296702dceef3ecad9589615ea52aa107047e8b1e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"m69m5lOMwpkz635qWoFERvFv3pI1ujjStht87Z9FyPPI8PsbpvIIJkxW33bjtYEOWS5B6+Uo/uMNiwLD1C5mBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-01T05:01:18.149747Z","bundle_sha256":"f3c885365b47a339cce2429da30cdab1d291420ec0006540e3980e72fea25000"}}