{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:YMRUUWA7SGVLUOXQKARTXCJT7A","short_pith_number":"pith:YMRUUWA7","canonical_record":{"source":{"id":"2406.13236","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-19T05:53:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d2c3a2af08fb96ec1d3567c998f710428420e4fd794f10f3dc9bf5d1321b7b01","abstract_canon_sha256":"3b2036fb67ff025a4b74870c98bd00c8f241f5969be988d18debfe082c41f635"},"schema_version":"1.0"},"canonical_sha256":"c3234a581f91aaba3af050233b8933f81b7ba6ff7ea5602be1f45ade4fee01ed","source":{"kind":"arxiv","id":"2406.13236","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.13236","created_at":"2026-07-05T09:28:32Z"},{"alias_kind":"arxiv_version","alias_value":"2406.13236v2","created_at":"2026-07-05T09:28:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.13236","created_at":"2026-07-05T09:28:32Z"},{"alias_kind":"pith_short_12","alias_value":"YMRUUWA7SGVL","created_at":"2026-07-05T09:28:32Z"},{"alias_kind":"pith_short_16","alias_value":"YMRUUWA7SGVLUOXQ","created_at":"2026-07-05T09:28:32Z"},{"alias_kind":"pith_short_8","alias_value":"YMRUUWA7","created_at":"2026-07-05T09:28:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:YMRUUWA7SGVLUOXQKARTXCJT7A","target":"record","payload":{"canonical_record":{"source":{"id":"2406.13236","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-19T05:53:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d2c3a2af08fb96ec1d3567c998f710428420e4fd794f10f3dc9bf5d1321b7b01","abstract_canon_sha256":"3b2036fb67ff025a4b74870c98bd00c8f241f5969be988d18debfe082c41f635"},"schema_version":"1.0"},"canonical_sha256":"c3234a581f91aaba3af050233b8933f81b7ba6ff7ea5602be1f45ade4fee01ed","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:28:32.598782Z","signature_b64":"PtWJNdQTQITO8bbRCzYS5rZ8bTjLDAmgW6gScdpVIR4e6kbFLPOunJgGAUUvAEmRvbpxv/Daj5ryLBJ5WldCAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c3234a581f91aaba3af050233b8933f81b7ba6ff7ea5602be1f45ade4fee01ed","last_reissued_at":"2026-07-05T09:28:32.598284Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:28:32.598284Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.13236","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":"2026-07-05T09:28:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Frig3JTS14RRn+b3IMB9R4iI9zscaw2+l+uMrGCUA+uPeOlMa3DZLZEVo3Bu3zXHUubfO24vY7NE8k7IYJmTBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T16:31:40.073334Z"},"content_sha256":"7838be162199215cd72423f12eac5e31ab2fc99714b5f5e1f2b48534f0476e54","schema_version":"1.0","event_id":"sha256:7838be162199215cd72423f12eac5e31ab2fc99714b5f5e1f2b48534f0476e54"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:YMRUUWA7SGVLUOXQKARTXCJT7A","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Data Contamination Can Cross Language Barriers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Animesh Kumar, Feng Yao, Jingbo Shang, Sunan Xu, Yufan Zhuang, Zihao Sun","submitted_at":"2024-06-19T05:53:27Z","abstract_excerpt":"The opacity in developing large language models (LLMs) is raising growing concerns about the potential contamination of public benchmarks in the pre-training data. Existing contamination detection methods are typically based on the text overlap between training and evaluation data, which can be too superficial to reflect deeper forms of contamination. In this paper, we first present a cross-lingual form of contamination that inflates LLMs' performance while evading current detection methods, deliberately injected by overfitting LLMs on the translated versions of benchmark test sets. Then, we p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.13236","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/2406.13236/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":"2026-07-05T09:28:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SHX1gQI4C8OhmKPrK2iAO9SoRHUEwRtKlCA05ZgxIzz3SQ+iChD3eyMJs6AHVYTHUIREBj0zuRkgVJIcPEGcDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T16:31:40.073966Z"},"content_sha256":"75d87c4a8c010d33cd44fd1832c740ce251f6c00ccdfc5d97388160e86f21d85","schema_version":"1.0","event_id":"sha256:75d87c4a8c010d33cd44fd1832c740ce251f6c00ccdfc5d97388160e86f21d85"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YMRUUWA7SGVLUOXQKARTXCJT7A/bundle.json","state_url":"https://pith.science/pith/YMRUUWA7SGVLUOXQKARTXCJT7A/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YMRUUWA7SGVLUOXQKARTXCJT7A/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":"2026-08-12T16:31:40Z","links":{"resolver":"https://pith.science/pith/YMRUUWA7SGVLUOXQKARTXCJT7A","bundle":"https://pith.science/pith/YMRUUWA7SGVLUOXQKARTXCJT7A/bundle.json","state":"https://pith.science/pith/YMRUUWA7SGVLUOXQKARTXCJT7A/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YMRUUWA7SGVLUOXQKARTXCJT7A/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:YMRUUWA7SGVLUOXQKARTXCJT7A","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":"3b2036fb67ff025a4b74870c98bd00c8f241f5969be988d18debfe082c41f635","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-19T05:53:27Z","title_canon_sha256":"d2c3a2af08fb96ec1d3567c998f710428420e4fd794f10f3dc9bf5d1321b7b01"},"schema_version":"1.0","source":{"id":"2406.13236","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.13236","created_at":"2026-07-05T09:28:32Z"},{"alias_kind":"arxiv_version","alias_value":"2406.13236v2","created_at":"2026-07-05T09:28:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.13236","created_at":"2026-07-05T09:28:32Z"},{"alias_kind":"pith_short_12","alias_value":"YMRUUWA7SGVL","created_at":"2026-07-05T09:28:32Z"},{"alias_kind":"pith_short_16","alias_value":"YMRUUWA7SGVLUOXQ","created_at":"2026-07-05T09:28:32Z"},{"alias_kind":"pith_short_8","alias_value":"YMRUUWA7","created_at":"2026-07-05T09:28:32Z"}],"graph_snapshots":[{"event_id":"sha256:75d87c4a8c010d33cd44fd1832c740ce251f6c00ccdfc5d97388160e86f21d85","target":"graph","created_at":"2026-07-05T09:28:32Z","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/2406.13236/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The opacity in developing large language models (LLMs) is raising growing concerns about the potential contamination of public benchmarks in the pre-training data. Existing contamination detection methods are typically based on the text overlap between training and evaluation data, which can be too superficial to reflect deeper forms of contamination. In this paper, we first present a cross-lingual form of contamination that inflates LLMs' performance while evading current detection methods, deliberately injected by overfitting LLMs on the translated versions of benchmark test sets. Then, we p","authors_text":"Animesh Kumar, Feng Yao, Jingbo Shang, Sunan Xu, Yufan Zhuang, Zihao Sun","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-19T05:53:27Z","title":"Data Contamination Can Cross Language Barriers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.13236","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":"sha256:7838be162199215cd72423f12eac5e31ab2fc99714b5f5e1f2b48534f0476e54","target":"record","created_at":"2026-07-05T09:28:32Z","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":"3b2036fb67ff025a4b74870c98bd00c8f241f5969be988d18debfe082c41f635","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-19T05:53:27Z","title_canon_sha256":"d2c3a2af08fb96ec1d3567c998f710428420e4fd794f10f3dc9bf5d1321b7b01"},"schema_version":"1.0","source":{"id":"2406.13236","kind":"arxiv","version":2}},"canonical_sha256":"c3234a581f91aaba3af050233b8933f81b7ba6ff7ea5602be1f45ade4fee01ed","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c3234a581f91aaba3af050233b8933f81b7ba6ff7ea5602be1f45ade4fee01ed","first_computed_at":"2026-07-05T09:28:32.598284Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:28:32.598284Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"PtWJNdQTQITO8bbRCzYS5rZ8bTjLDAmgW6gScdpVIR4e6kbFLPOunJgGAUUvAEmRvbpxv/Daj5ryLBJ5WldCAg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:28:32.598782Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.13236","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7838be162199215cd72423f12eac5e31ab2fc99714b5f5e1f2b48534f0476e54","sha256:75d87c4a8c010d33cd44fd1832c740ce251f6c00ccdfc5d97388160e86f21d85"],"state_sha256":"d31e051d344764078cbfb5ac747b69454fd1a4b5ed00a4cd0fb48455acf006a3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Lw/KcPaghwQSeiZt6zJXGagXwWLgI3nmjlL6QqtRAzvGkFylewTgzHbC1bLlimrnYcRd73NiEfJHhLzuaiJtBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-12T16:31:40.079878Z","bundle_sha256":"9c7dda88a4e0ea9922fc8101f7b19035f73c68116fc3749eeb8f9404cb270bc7"}}