{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:VUH64C5CCTASECKDZSQSV3VHPC","short_pith_number":"pith:VUH64C5C","canonical_record":{"source":{"id":"2508.13180","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-12T22:52:21Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"1387dd8f9c643006e32a2c239e0ec69633a81c584a566fa59505483d30f65568","abstract_canon_sha256":"cb1e0ce03096c6685ab33559e22e90dcec5b4eac121923579c14a3946416ca25"},"schema_version":"1.0"},"canonical_sha256":"ad0fee0ba214c1220943cca12aeea778bce6232874ffd464795a0159f66ab8a1","source":{"kind":"arxiv","id":"2508.13180","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.13180","created_at":"2026-07-05T11:55:43Z"},{"alias_kind":"arxiv_version","alias_value":"2508.13180v1","created_at":"2026-07-05T11:55:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.13180","created_at":"2026-07-05T11:55:43Z"},{"alias_kind":"pith_short_12","alias_value":"VUH64C5CCTAS","created_at":"2026-07-05T11:55:43Z"},{"alias_kind":"pith_short_16","alias_value":"VUH64C5CCTASECKD","created_at":"2026-07-05T11:55:43Z"},{"alias_kind":"pith_short_8","alias_value":"VUH64C5C","created_at":"2026-07-05T11:55:43Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:VUH64C5CCTASECKDZSQSV3VHPC","target":"record","payload":{"canonical_record":{"source":{"id":"2508.13180","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-12T22:52:21Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"1387dd8f9c643006e32a2c239e0ec69633a81c584a566fa59505483d30f65568","abstract_canon_sha256":"cb1e0ce03096c6685ab33559e22e90dcec5b4eac121923579c14a3946416ca25"},"schema_version":"1.0"},"canonical_sha256":"ad0fee0ba214c1220943cca12aeea778bce6232874ffd464795a0159f66ab8a1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:55:43.811465Z","signature_b64":"fQcd1MnZzUcU48OA2HWlIZY9oJWckxXJQKx2hdbdql14t5Fb4lDpdML5QnnE5ajVz9xbFwIytDJEOYYmkKcKDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ad0fee0ba214c1220943cca12aeea778bce6232874ffd464795a0159f66ab8a1","last_reissued_at":"2026-07-05T11:55:43.810988Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:55:43.810988Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.13180","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":"2026-07-05T11:55:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aIcZ+rvBSMRLt/jlSrERIj/7UWP61JNIkMRZ3+aq/Rn5ljb0zO8LO9N7zuiuIZo80BpPj35opXWjXMlA8OflDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-12T15:23:34.322879Z"},"content_sha256":"dc1f851dd91d3b05f17a9e7ee9eb89ab852de47603e9ecc60c6e19cd939cad65","schema_version":"1.0","event_id":"sha256:dc1f851dd91d3b05f17a9e7ee9eb89ab852de47603e9ecc60c6e19cd939cad65"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:VUH64C5CCTASECKDZSQSV3VHPC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Search-Time Data Contamination","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Julian Michael, Meher Mankikar, Zifan Wang, Ziwen Han","submitted_at":"2025-08-12T22:52:21Z","abstract_excerpt":"Data contamination refers to the leakage of evaluation data into model training data, resulting in overfitting to supposedly held-out test sets and compromising test validity. We identify an analogous issue, search-time contamination (STC), in evaluating search-based LLM agents which use tools to gather information from online sources when answering user queries. STC occurs when the retrieval step surfaces a source containing the test question (or a near-duplicate) alongside its answer, enabling agents to copy rather than genuinely infer or reason, undermining benchmark integrity. We find that"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.13180","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/2508.13180/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":"2026-07-05T11:55:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xE1FSJSyFrg7MEajLhztT208BG4ZtmRJq7d/MPqhk4cpT4pDgwyeiL8BXjKuIec8z9134vN0GRIU8reAAdKxAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-12T15:23:34.323286Z"},"content_sha256":"7947f0e77159da84dcd525d072eea1ceb0e6f6c6ad8603ee04c9b0b029214c66","schema_version":"1.0","event_id":"sha256:7947f0e77159da84dcd525d072eea1ceb0e6f6c6ad8603ee04c9b0b029214c66"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VUH64C5CCTASECKDZSQSV3VHPC/bundle.json","state_url":"https://pith.science/pith/VUH64C5CCTASECKDZSQSV3VHPC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VUH64C5CCTASECKDZSQSV3VHPC/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":"2026-07-12T15:23:34Z","links":{"resolver":"https://pith.science/pith/VUH64C5CCTASECKDZSQSV3VHPC","bundle":"https://pith.science/pith/VUH64C5CCTASECKDZSQSV3VHPC/bundle.json","state":"https://pith.science/pith/VUH64C5CCTASECKDZSQSV3VHPC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VUH64C5CCTASECKDZSQSV3VHPC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:VUH64C5CCTASECKDZSQSV3VHPC","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":"cb1e0ce03096c6685ab33559e22e90dcec5b4eac121923579c14a3946416ca25","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-12T22:52:21Z","title_canon_sha256":"1387dd8f9c643006e32a2c239e0ec69633a81c584a566fa59505483d30f65568"},"schema_version":"1.0","source":{"id":"2508.13180","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.13180","created_at":"2026-07-05T11:55:43Z"},{"alias_kind":"arxiv_version","alias_value":"2508.13180v1","created_at":"2026-07-05T11:55:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.13180","created_at":"2026-07-05T11:55:43Z"},{"alias_kind":"pith_short_12","alias_value":"VUH64C5CCTAS","created_at":"2026-07-05T11:55:43Z"},{"alias_kind":"pith_short_16","alias_value":"VUH64C5CCTASECKD","created_at":"2026-07-05T11:55:43Z"},{"alias_kind":"pith_short_8","alias_value":"VUH64C5C","created_at":"2026-07-05T11:55:43Z"}],"graph_snapshots":[{"event_id":"sha256:7947f0e77159da84dcd525d072eea1ceb0e6f6c6ad8603ee04c9b0b029214c66","target":"graph","created_at":"2026-07-05T11:55:43Z","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/2508.13180/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Data contamination refers to the leakage of evaluation data into model training data, resulting in overfitting to supposedly held-out test sets and compromising test validity. We identify an analogous issue, search-time contamination (STC), in evaluating search-based LLM agents which use tools to gather information from online sources when answering user queries. STC occurs when the retrieval step surfaces a source containing the test question (or a near-duplicate) alongside its answer, enabling agents to copy rather than genuinely infer or reason, undermining benchmark integrity. We find that","authors_text":"Julian Michael, Meher Mankikar, Zifan Wang, Ziwen Han","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-12T22:52:21Z","title":"Search-Time Data Contamination"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.13180","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:dc1f851dd91d3b05f17a9e7ee9eb89ab852de47603e9ecc60c6e19cd939cad65","target":"record","created_at":"2026-07-05T11:55:43Z","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":"cb1e0ce03096c6685ab33559e22e90dcec5b4eac121923579c14a3946416ca25","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-12T22:52:21Z","title_canon_sha256":"1387dd8f9c643006e32a2c239e0ec69633a81c584a566fa59505483d30f65568"},"schema_version":"1.0","source":{"id":"2508.13180","kind":"arxiv","version":1}},"canonical_sha256":"ad0fee0ba214c1220943cca12aeea778bce6232874ffd464795a0159f66ab8a1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ad0fee0ba214c1220943cca12aeea778bce6232874ffd464795a0159f66ab8a1","first_computed_at":"2026-07-05T11:55:43.810988Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:55:43.810988Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fQcd1MnZzUcU48OA2HWlIZY9oJWckxXJQKx2hdbdql14t5Fb4lDpdML5QnnE5ajVz9xbFwIytDJEOYYmkKcKDw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:55:43.811465Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.13180","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dc1f851dd91d3b05f17a9e7ee9eb89ab852de47603e9ecc60c6e19cd939cad65","sha256:7947f0e77159da84dcd525d072eea1ceb0e6f6c6ad8603ee04c9b0b029214c66"],"state_sha256":"e0b5fc5e0d13d2e9650e85d424bd04e23f52458cc8418a0d9b20c31b28142d66"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jrJ8mBR63D+dW4YmJQ/TL6x4UG8ei5RCIqiWXUg3ykXYY9+hq8YpErAW8v6biJp0JvZhSgDqwjIDEpkLFActCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-12T15:23:34.326100Z","bundle_sha256":"6474915071ae59c149b97f5f3540f37b73764cc86bd562f2ecdcaa521beab980"}}