{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:KXKXSW6QMMRCLCE5ACS2SMWUSL","short_pith_number":"pith:KXKXSW6Q","canonical_record":{"source":{"id":"2401.04621","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-01-09T15:46:38Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"4811fa6ce5d5620032edf26dfc9fee63a27984df427eb4662569b626ee660ad7","abstract_canon_sha256":"af71da4e606047ade0f2aa99f33231aa8be3494171037132ea84eda2ae962b06"},"schema_version":"1.0"},"canonical_sha256":"55d5795bd0632225889d00a5a932d492db831f1f12f470d32d99da2549107427","source":{"kind":"arxiv","id":"2401.04621","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.04621","created_at":"2026-07-05T08:28:04Z"},{"alias_kind":"arxiv_version","alias_value":"2401.04621v3","created_at":"2026-07-05T08:28:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.04621","created_at":"2026-07-05T08:28:04Z"},{"alias_kind":"pith_short_12","alias_value":"KXKXSW6QMMRC","created_at":"2026-07-05T08:28:04Z"},{"alias_kind":"pith_short_16","alias_value":"KXKXSW6QMMRCLCE5","created_at":"2026-07-05T08:28:04Z"},{"alias_kind":"pith_short_8","alias_value":"KXKXSW6Q","created_at":"2026-07-05T08:28:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:KXKXSW6QMMRCLCE5ACS2SMWUSL","target":"record","payload":{"canonical_record":{"source":{"id":"2401.04621","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-01-09T15:46:38Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"4811fa6ce5d5620032edf26dfc9fee63a27984df427eb4662569b626ee660ad7","abstract_canon_sha256":"af71da4e606047ade0f2aa99f33231aa8be3494171037132ea84eda2ae962b06"},"schema_version":"1.0"},"canonical_sha256":"55d5795bd0632225889d00a5a932d492db831f1f12f470d32d99da2549107427","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:28:04.739959Z","signature_b64":"IYJC5yQVGgSL0EtR61Wkmr5UDX2Jpf8eW2rlEitHvspKvr+B25rNaqxlUDXYRrcgyVcmkMyBh98XfmXWA9/MBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"55d5795bd0632225889d00a5a932d492db831f1f12f470d32d99da2549107427","last_reissued_at":"2026-07-05T08:28:04.739514Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:28:04.739514Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2401.04621","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:28:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MZ97d1Ucw+qXjtJ+pCimZaBghIKxgSR5KmeUdk9FxpqnZviJkOPCvhMDr68jbWOYIMmmu1p3HBZQoQSKlCtMDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T14:06:08.078662Z"},"content_sha256":"e97f630df0b7175f90dbdf647444ded5c6e0e43f0ad93f1212afa5f84b23c198","schema_version":"1.0","event_id":"sha256:e97f630df0b7175f90dbdf647444ded5c6e0e43f0ad93f1212afa5f84b23c198"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:KXKXSW6QMMRCLCE5ACS2SMWUSL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"DebugBench: Evaluating Debugging Capability of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.SE","authors_text":"Haotian Hui, Maosong Sun, Runchu Tian, Weichuan Liu, Xin Cong, Yankai Lin, Yesai Wu, Yining Ye, Yinxu Pan, Yujia Qin, Zhiyuan Liu","submitted_at":"2024-01-09T15:46:38Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated exceptional coding capability. However, as another critical component of programming proficiency, the debugging capability of LLMs remains relatively unexplored. Previous evaluations of LLMs' debugging ability are significantly limited by the risk of data leakage, the scale of the dataset, and the variety of tested bugs. To overcome these deficiencies, we introduce `DebugBench', an LLM debugging benchmark consisting of 4,253 instances. It covers four major bug categories and 18 minor types in C++, Java, and Python. To construct DebugBench, we coll"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.04621","kind":"arxiv","version":3},"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/2401.04621/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":"2026-07-05T08:28:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vQUSMytiB7plWIIc3Qr/nKAXKLCTEDO4XjTFJw+5aQFJk9TFcIc68SJxueC4BTWD6aSzhQI4jm318uaCdVLrCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T14:06:08.079041Z"},"content_sha256":"961bb646d6c266911898641682a2d411e7b749ab25f1dacc935799e6da4cea11","schema_version":"1.0","event_id":"sha256:961bb646d6c266911898641682a2d411e7b749ab25f1dacc935799e6da4cea11"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KXKXSW6QMMRCLCE5ACS2SMWUSL/bundle.json","state_url":"https://pith.science/pith/KXKXSW6QMMRCLCE5ACS2SMWUSL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KXKXSW6QMMRCLCE5ACS2SMWUSL/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":"2026-07-06T14:06:08Z","links":{"resolver":"https://pith.science/pith/KXKXSW6QMMRCLCE5ACS2SMWUSL","bundle":"https://pith.science/pith/KXKXSW6QMMRCLCE5ACS2SMWUSL/bundle.json","state":"https://pith.science/pith/KXKXSW6QMMRCLCE5ACS2SMWUSL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KXKXSW6QMMRCLCE5ACS2SMWUSL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:KXKXSW6QMMRCLCE5ACS2SMWUSL","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":"af71da4e606047ade0f2aa99f33231aa8be3494171037132ea84eda2ae962b06","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-01-09T15:46:38Z","title_canon_sha256":"4811fa6ce5d5620032edf26dfc9fee63a27984df427eb4662569b626ee660ad7"},"schema_version":"1.0","source":{"id":"2401.04621","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.04621","created_at":"2026-07-05T08:28:04Z"},{"alias_kind":"arxiv_version","alias_value":"2401.04621v3","created_at":"2026-07-05T08:28:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.04621","created_at":"2026-07-05T08:28:04Z"},{"alias_kind":"pith_short_12","alias_value":"KXKXSW6QMMRC","created_at":"2026-07-05T08:28:04Z"},{"alias_kind":"pith_short_16","alias_value":"KXKXSW6QMMRCLCE5","created_at":"2026-07-05T08:28:04Z"},{"alias_kind":"pith_short_8","alias_value":"KXKXSW6Q","created_at":"2026-07-05T08:28:04Z"}],"graph_snapshots":[{"event_id":"sha256:961bb646d6c266911898641682a2d411e7b749ab25f1dacc935799e6da4cea11","target":"graph","created_at":"2026-07-05T08:28:04Z","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/2401.04621/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have demonstrated exceptional coding capability. However, as another critical component of programming proficiency, the debugging capability of LLMs remains relatively unexplored. Previous evaluations of LLMs' debugging ability are significantly limited by the risk of data leakage, the scale of the dataset, and the variety of tested bugs. To overcome these deficiencies, we introduce `DebugBench', an LLM debugging benchmark consisting of 4,253 instances. It covers four major bug categories and 18 minor types in C++, Java, and Python. To construct DebugBench, we coll","authors_text":"Haotian Hui, Maosong Sun, Runchu Tian, Weichuan Liu, Xin Cong, Yankai Lin, Yesai Wu, Yining Ye, Yinxu Pan, Yujia Qin, Zhiyuan Liu","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-01-09T15:46:38Z","title":"DebugBench: Evaluating Debugging Capability of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.04621","kind":"arxiv","version":3},"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:e97f630df0b7175f90dbdf647444ded5c6e0e43f0ad93f1212afa5f84b23c198","target":"record","created_at":"2026-07-05T08:28:04Z","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":"af71da4e606047ade0f2aa99f33231aa8be3494171037132ea84eda2ae962b06","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-01-09T15:46:38Z","title_canon_sha256":"4811fa6ce5d5620032edf26dfc9fee63a27984df427eb4662569b626ee660ad7"},"schema_version":"1.0","source":{"id":"2401.04621","kind":"arxiv","version":3}},"canonical_sha256":"55d5795bd0632225889d00a5a932d492db831f1f12f470d32d99da2549107427","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"55d5795bd0632225889d00a5a932d492db831f1f12f470d32d99da2549107427","first_computed_at":"2026-07-05T08:28:04.739514Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:28:04.739514Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"IYJC5yQVGgSL0EtR61Wkmr5UDX2Jpf8eW2rlEitHvspKvr+B25rNaqxlUDXYRrcgyVcmkMyBh98XfmXWA9/MBw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:28:04.739959Z","signed_message":"canonical_sha256_bytes"},"source_id":"2401.04621","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e97f630df0b7175f90dbdf647444ded5c6e0e43f0ad93f1212afa5f84b23c198","sha256:961bb646d6c266911898641682a2d411e7b749ab25f1dacc935799e6da4cea11"],"state_sha256":"1ac9ec67383d5c61a606c5c7aab799b5f44cdba37b55f7c62047f08cea071194"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PGQTW6JLzTjcgHQeoqJjt9TrQ37evJSYG2XAMal2W8laNi9h2PSQ3ucC7q1iKWo32uz2P/HwJRXKtLnGtDVCDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T14:06:08.080975Z","bundle_sha256":"271f20de79975bb83b61f0a0e84a4c8ce53a66476eaccd6b3b1c746432a727b8"}}