{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:Q6LPEAMDYTVMDFIZ5LF267CSMG","short_pith_number":"pith:Q6LPEAMD","canonical_record":{"source":{"id":"2404.13076","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-15T16:49:59Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"515f95f1cdfd1496b2c014e6f5952537a265a6235804636bfa7ef96562515622","abstract_canon_sha256":"ffebaa14a97d3d1ef8a2cfc5eaa551b244bacb5b400e5cd720bb46621ebe2b36"},"schema_version":"1.0"},"canonical_sha256":"8796f20183c4eac19519eacbaf7c5261ab3dc8dba448bc1bd90f5875424e434a","source":{"kind":"arxiv","id":"2404.13076","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.13076","created_at":"2026-05-22T18:40:04Z"},{"alias_kind":"arxiv_version","alias_value":"2404.13076v1","created_at":"2026-05-22T18:40:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.13076","created_at":"2026-05-22T18:40:04Z"},{"alias_kind":"pith_short_12","alias_value":"Q6LPEAMDYTVM","created_at":"2026-05-22T18:40:04Z"},{"alias_kind":"pith_short_16","alias_value":"Q6LPEAMDYTVMDFIZ","created_at":"2026-05-22T18:40:04Z"},{"alias_kind":"pith_short_8","alias_value":"Q6LPEAMD","created_at":"2026-05-22T18:40:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:Q6LPEAMDYTVMDFIZ5LF267CSMG","target":"record","payload":{"canonical_record":{"source":{"id":"2404.13076","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-15T16:49:59Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"515f95f1cdfd1496b2c014e6f5952537a265a6235804636bfa7ef96562515622","abstract_canon_sha256":"ffebaa14a97d3d1ef8a2cfc5eaa551b244bacb5b400e5cd720bb46621ebe2b36"},"schema_version":"1.0"},"canonical_sha256":"8796f20183c4eac19519eacbaf7c5261ab3dc8dba448bc1bd90f5875424e434a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T18:40:04.775271Z","signature_b64":"hLruYd3j2QUvJmpNHj6+4cGzaktePUaSBVzlCgN9g7W1eKO183SDqUqgL3vyTIdYFpzr0pYcKMw1VR2pWtxPCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8796f20183c4eac19519eacbaf7c5261ab3dc8dba448bc1bd90f5875424e434a","last_reissued_at":"2026-05-22T18:40:04.772304Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T18:40:04.772304Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2404.13076","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":"2026-05-22T18:40:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"j/oyVSt367Ldd4dP9z+c4GjTmKDg8UIM4+y3hdurSrEvl4pNM7Gw9dGuNQDA7wlHhN8cmw+bvygEoRRRFtMeBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-23T05:02:00.839512Z"},"content_sha256":"36f3bad9cfb1fa5306b2d5cc5359e9eb6d3628ae49f78871ca4a606c5444a962","schema_version":"1.0","event_id":"sha256:36f3bad9cfb1fa5306b2d5cc5359e9eb6d3628ae49f78871ca4a606c5444a962"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:Q6LPEAMDYTVMDFIZ5LF267CSMG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LLM Evaluators Recognize and Favor Their Own Generations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Arjun Panickssery, Samuel R. Bowman, Shi Feng","submitted_at":"2024-04-15T16:49:59Z","abstract_excerpt":"Self-evaluation using large language models (LLMs) has proven valuable not only in benchmarking but also methods like reward modeling, constitutional AI, and self-refinement. But new biases are introduced due to the same LLM acting as both the evaluator and the evaluatee. One such bias is self-preference, where an LLM evaluator scores its own outputs higher than others' while human annotators consider them of equal quality. But do LLMs actually recognize their own outputs when they give those texts higher scores, or is it just a coincidence? In this paper, we investigate if self-recognition ca"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.13076","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/2404.13076/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":"2026-05-22T18:40:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NIo72VbXOUAijitG1MP9O5d/3zE8gBzjk2TfxnTy3fl4NNh1Xua2HE8x97JhD0zbLVK1Tla7w2PVEQUqegUyAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-23T05:02:00.840226Z"},"content_sha256":"41b4e8356f616105350c7a1e602b73334ca6a274f881c7354e2ea1026add9e18","schema_version":"1.0","event_id":"sha256:41b4e8356f616105350c7a1e602b73334ca6a274f881c7354e2ea1026add9e18"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Q6LPEAMDYTVMDFIZ5LF267CSMG/bundle.json","state_url":"https://pith.science/pith/Q6LPEAMDYTVMDFIZ5LF267CSMG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Q6LPEAMDYTVMDFIZ5LF267CSMG/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":"2026-05-23T05:02:00Z","links":{"resolver":"https://pith.science/pith/Q6LPEAMDYTVMDFIZ5LF267CSMG","bundle":"https://pith.science/pith/Q6LPEAMDYTVMDFIZ5LF267CSMG/bundle.json","state":"https://pith.science/pith/Q6LPEAMDYTVMDFIZ5LF267CSMG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Q6LPEAMDYTVMDFIZ5LF267CSMG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:Q6LPEAMDYTVMDFIZ5LF267CSMG","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":"ffebaa14a97d3d1ef8a2cfc5eaa551b244bacb5b400e5cd720bb46621ebe2b36","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-15T16:49:59Z","title_canon_sha256":"515f95f1cdfd1496b2c014e6f5952537a265a6235804636bfa7ef96562515622"},"schema_version":"1.0","source":{"id":"2404.13076","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.13076","created_at":"2026-05-22T18:40:04Z"},{"alias_kind":"arxiv_version","alias_value":"2404.13076v1","created_at":"2026-05-22T18:40:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.13076","created_at":"2026-05-22T18:40:04Z"},{"alias_kind":"pith_short_12","alias_value":"Q6LPEAMDYTVM","created_at":"2026-05-22T18:40:04Z"},{"alias_kind":"pith_short_16","alias_value":"Q6LPEAMDYTVMDFIZ","created_at":"2026-05-22T18:40:04Z"},{"alias_kind":"pith_short_8","alias_value":"Q6LPEAMD","created_at":"2026-05-22T18:40:04Z"}],"graph_snapshots":[{"event_id":"sha256:41b4e8356f616105350c7a1e602b73334ca6a274f881c7354e2ea1026add9e18","target":"graph","created_at":"2026-05-22T18:40:04Z","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/2404.13076/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Self-evaluation using large language models (LLMs) has proven valuable not only in benchmarking but also methods like reward modeling, constitutional AI, and self-refinement. But new biases are introduced due to the same LLM acting as both the evaluator and the evaluatee. One such bias is self-preference, where an LLM evaluator scores its own outputs higher than others' while human annotators consider them of equal quality. But do LLMs actually recognize their own outputs when they give those texts higher scores, or is it just a coincidence? In this paper, we investigate if self-recognition ca","authors_text":"Arjun Panickssery, Samuel R. Bowman, Shi Feng","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-15T16:49:59Z","title":"LLM Evaluators Recognize and Favor Their Own Generations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.13076","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":"sha256:36f3bad9cfb1fa5306b2d5cc5359e9eb6d3628ae49f78871ca4a606c5444a962","target":"record","created_at":"2026-05-22T18:40:04Z","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":"ffebaa14a97d3d1ef8a2cfc5eaa551b244bacb5b400e5cd720bb46621ebe2b36","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-15T16:49:59Z","title_canon_sha256":"515f95f1cdfd1496b2c014e6f5952537a265a6235804636bfa7ef96562515622"},"schema_version":"1.0","source":{"id":"2404.13076","kind":"arxiv","version":1}},"canonical_sha256":"8796f20183c4eac19519eacbaf7c5261ab3dc8dba448bc1bd90f5875424e434a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8796f20183c4eac19519eacbaf7c5261ab3dc8dba448bc1bd90f5875424e434a","first_computed_at":"2026-05-22T18:40:04.772304Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-22T18:40:04.772304Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hLruYd3j2QUvJmpNHj6+4cGzaktePUaSBVzlCgN9g7W1eKO183SDqUqgL3vyTIdYFpzr0pYcKMw1VR2pWtxPCA==","signature_status":"signed_v1","signed_at":"2026-05-22T18:40:04.775271Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.13076","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:36f3bad9cfb1fa5306b2d5cc5359e9eb6d3628ae49f78871ca4a606c5444a962","sha256:41b4e8356f616105350c7a1e602b73334ca6a274f881c7354e2ea1026add9e18"],"state_sha256":"5edb94b697c18c42420fe0aea2629e7488673c3791fdfa5820730597e9b5293e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YWylYIdpbfEhlT0URExdDysyilAuDlvvGpbBPqEQYgpFBoI0FLEgUzraR7u2n70Ajdy0QFZo9FOhBHPs5TYADw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-23T05:02:00.843790Z","bundle_sha256":"e9be15d4159ff87beba6a4f795e432b45010ad1ffcd4526737abb1994e71ce35"}}