{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:QAR5MQYLBXCQYNEHM5KGHJSF4P","short_pith_number":"pith:QAR5MQYL","canonical_record":{"source":{"id":"2606.23313","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2026-06-22T13:26:29Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"ea901f8831d1b5244f7a56e8d77ee85e5c6ff3faa25edcc16828d39e45e2ec98","abstract_canon_sha256":"764ae675698d41d43cacb314f65c1594ccdca5c3c3c942d2ff62210586b68316"},"schema_version":"1.0"},"canonical_sha256":"8023d6430b0dc50c3487675463a645e3d66f8021ce2e2426ce396b056141ea89","source":{"kind":"arxiv","id":"2606.23313","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.23313","created_at":"2026-06-23T03:14:16Z"},{"alias_kind":"arxiv_version","alias_value":"2606.23313v1","created_at":"2026-06-23T03:14:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.23313","created_at":"2026-06-23T03:14:16Z"},{"alias_kind":"pith_short_12","alias_value":"QAR5MQYLBXCQ","created_at":"2026-06-23T03:14:16Z"},{"alias_kind":"pith_short_16","alias_value":"QAR5MQYLBXCQYNEH","created_at":"2026-06-23T03:14:16Z"},{"alias_kind":"pith_short_8","alias_value":"QAR5MQYL","created_at":"2026-06-23T03:14:16Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:QAR5MQYLBXCQYNEHM5KGHJSF4P","target":"record","payload":{"canonical_record":{"source":{"id":"2606.23313","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2026-06-22T13:26:29Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"ea901f8831d1b5244f7a56e8d77ee85e5c6ff3faa25edcc16828d39e45e2ec98","abstract_canon_sha256":"764ae675698d41d43cacb314f65c1594ccdca5c3c3c942d2ff62210586b68316"},"schema_version":"1.0"},"canonical_sha256":"8023d6430b0dc50c3487675463a645e3d66f8021ce2e2426ce396b056141ea89","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T03:14:16.604403Z","signature_b64":"unUYqPDznkZauIHNHWS5kCKbdk4u0hxmD1b7HezW51W5EnSTwzyq1q3xN/WKqHmjMvDIAl6DXeuRPn0hnl/XAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8023d6430b0dc50c3487675463a645e3d66f8021ce2e2426ce396b056141ea89","last_reissued_at":"2026-06-23T03:14:16.604060Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T03:14:16.604060Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.23313","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":"2026-06-23T03:14:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Q/X+5WhCNtUA28P3pZOxiaONeCy7V1bQKBZTpSG5T9ZzYMHKXPkV2Y7V2dZQNqp/8C2HJZCGW2Uwow0pDfDsAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-28T18:09:13.046798Z"},"content_sha256":"7c551ae591cd150588054b99ad1b97025ab48b11e40b1668bebcfe686fb00f45","schema_version":"1.0","event_id":"sha256:7c551ae591cd150588054b99ad1b97025ab48b11e40b1668bebcfe686fb00f45"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:QAR5MQYLBXCQYNEHM5KGHJSF4P","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Uncertainty-based Debiasing and Unlearning for Decontamination","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CY","authors_text":"Guangzhi Sun, Mark Gales, Xiao Zhan","submitted_at":"2026-06-22T13:26:29Z","abstract_excerpt":"Benchmark-based evaluation is the dominant paradigm for assessing large language model (LLM) capabilities, yet data contamination inflates reported performance and undermines fair comparison. Existing decontamination methods are evaluated solely through aggregate accuracy, which can obscure substantial differences in per-sample model behaviour, and many require access to an uncontaminated model. In this paper, we propose a sample-level evaluation framework for decontamination that complements accuracy-based assessment with distributional distance metrics, measuring how closely a decontaminated"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.23313","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/2606.23313/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":"2026-06-23T03:14:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xa+MFEQ3h1wr6ohAyz4DUeB+igMab80OY7+Lz9kfU7ZWuczGSa2X5z0xEfPvVfj5Pr6iM71e2fT8jap/3RTWAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-28T18:09:13.047400Z"},"content_sha256":"1b4171de5f4f5be7bb8a115b2215de90465c4049a84343625730d0f28250474b","schema_version":"1.0","event_id":"sha256:1b4171de5f4f5be7bb8a115b2215de90465c4049a84343625730d0f28250474b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QAR5MQYLBXCQYNEHM5KGHJSF4P/bundle.json","state_url":"https://pith.science/pith/QAR5MQYLBXCQYNEHM5KGHJSF4P/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QAR5MQYLBXCQYNEHM5KGHJSF4P/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":"2026-06-28T18:09:13Z","links":{"resolver":"https://pith.science/pith/QAR5MQYLBXCQYNEHM5KGHJSF4P","bundle":"https://pith.science/pith/QAR5MQYLBXCQYNEHM5KGHJSF4P/bundle.json","state":"https://pith.science/pith/QAR5MQYLBXCQYNEHM5KGHJSF4P/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QAR5MQYLBXCQYNEHM5KGHJSF4P/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:QAR5MQYLBXCQYNEHM5KGHJSF4P","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":"764ae675698d41d43cacb314f65c1594ccdca5c3c3c942d2ff62210586b68316","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2026-06-22T13:26:29Z","title_canon_sha256":"ea901f8831d1b5244f7a56e8d77ee85e5c6ff3faa25edcc16828d39e45e2ec98"},"schema_version":"1.0","source":{"id":"2606.23313","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.23313","created_at":"2026-06-23T03:14:16Z"},{"alias_kind":"arxiv_version","alias_value":"2606.23313v1","created_at":"2026-06-23T03:14:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.23313","created_at":"2026-06-23T03:14:16Z"},{"alias_kind":"pith_short_12","alias_value":"QAR5MQYLBXCQ","created_at":"2026-06-23T03:14:16Z"},{"alias_kind":"pith_short_16","alias_value":"QAR5MQYLBXCQYNEH","created_at":"2026-06-23T03:14:16Z"},{"alias_kind":"pith_short_8","alias_value":"QAR5MQYL","created_at":"2026-06-23T03:14:16Z"}],"graph_snapshots":[{"event_id":"sha256:1b4171de5f4f5be7bb8a115b2215de90465c4049a84343625730d0f28250474b","target":"graph","created_at":"2026-06-23T03:14:16Z","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/2606.23313/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Benchmark-based evaluation is the dominant paradigm for assessing large language model (LLM) capabilities, yet data contamination inflates reported performance and undermines fair comparison. Existing decontamination methods are evaluated solely through aggregate accuracy, which can obscure substantial differences in per-sample model behaviour, and many require access to an uncontaminated model. In this paper, we propose a sample-level evaluation framework for decontamination that complements accuracy-based assessment with distributional distance metrics, measuring how closely a decontaminated","authors_text":"Guangzhi Sun, Mark Gales, Xiao Zhan","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2026-06-22T13:26:29Z","title":"Uncertainty-based Debiasing and Unlearning for Decontamination"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.23313","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":"sha256:7c551ae591cd150588054b99ad1b97025ab48b11e40b1668bebcfe686fb00f45","target":"record","created_at":"2026-06-23T03:14:16Z","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":"764ae675698d41d43cacb314f65c1594ccdca5c3c3c942d2ff62210586b68316","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2026-06-22T13:26:29Z","title_canon_sha256":"ea901f8831d1b5244f7a56e8d77ee85e5c6ff3faa25edcc16828d39e45e2ec98"},"schema_version":"1.0","source":{"id":"2606.23313","kind":"arxiv","version":1}},"canonical_sha256":"8023d6430b0dc50c3487675463a645e3d66f8021ce2e2426ce396b056141ea89","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8023d6430b0dc50c3487675463a645e3d66f8021ce2e2426ce396b056141ea89","first_computed_at":"2026-06-23T03:14:16.604060Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-23T03:14:16.604060Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"unUYqPDznkZauIHNHWS5kCKbdk4u0hxmD1b7HezW51W5EnSTwzyq1q3xN/WKqHmjMvDIAl6DXeuRPn0hnl/XAg==","signature_status":"signed_v1","signed_at":"2026-06-23T03:14:16.604403Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.23313","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7c551ae591cd150588054b99ad1b97025ab48b11e40b1668bebcfe686fb00f45","sha256:1b4171de5f4f5be7bb8a115b2215de90465c4049a84343625730d0f28250474b"],"state_sha256":"b8415ba641efa61b124dc1b7d9b5c2dfdf8191f455bda5bc518de27b4ec182c5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CKwJzYyPyhcaGSrpTpQfT1Axl44rXNvve6ASgS/YN9Rc0bAclIOVrrkMrhCAlH8R8Qjj+GmLHmZShaxMlFSvBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-28T18:09:13.050056Z","bundle_sha256":"f5b38ce562a12391a86f37b05cdac22ec3a9df56515882abb92797af3f557539"}}