{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:YLAUA24JES6YV253KVP5FLQ43N","short_pith_number":"pith:YLAUA24J","canonical_record":{"source":{"id":"2508.06709","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-08T21:22:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"883df8c3e2ec65e16541d809357732cd712c0dc41c85ac87c67dae6b36e12bb6","abstract_canon_sha256":"f3b3979d516d168db47d86b63501621e7db8d5393025a0716ff81ebf7701a8f6"},"schema_version":"1.0"},"canonical_sha256":"c2c1406b8924bd8aebbb555fd2ae1cdb56c26663272e463e152f88fd441e261d","source":{"kind":"arxiv","id":"2508.06709","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.06709","created_at":"2026-07-05T11:51:05Z"},{"alias_kind":"arxiv_version","alias_value":"2508.06709v1","created_at":"2026-07-05T11:51:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.06709","created_at":"2026-07-05T11:51:05Z"},{"alias_kind":"pith_short_12","alias_value":"YLAUA24JES6Y","created_at":"2026-07-05T11:51:05Z"},{"alias_kind":"pith_short_16","alias_value":"YLAUA24JES6YV253","created_at":"2026-07-05T11:51:05Z"},{"alias_kind":"pith_short_8","alias_value":"YLAUA24J","created_at":"2026-07-05T11:51:05Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:YLAUA24JES6YV253KVP5FLQ43N","target":"record","payload":{"canonical_record":{"source":{"id":"2508.06709","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-08T21:22:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"883df8c3e2ec65e16541d809357732cd712c0dc41c85ac87c67dae6b36e12bb6","abstract_canon_sha256":"f3b3979d516d168db47d86b63501621e7db8d5393025a0716ff81ebf7701a8f6"},"schema_version":"1.0"},"canonical_sha256":"c2c1406b8924bd8aebbb555fd2ae1cdb56c26663272e463e152f88fd441e261d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:51:05.314043Z","signature_b64":"n9d+NOo7IxwzbhHeo3ZDXuIHcFo2xfRaks3gp7MW+z6a+Hq0GIpxbFokdo/0M53nKrvD72PA211Y3tDwFqzcBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c2c1406b8924bd8aebbb555fd2ae1cdb56c26663272e463e152f88fd441e261d","last_reissued_at":"2026-07-05T11:51:05.313415Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:51:05.313415Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.06709","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":"2026-07-05T11:51:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XCC62orNqdRscpBGe7j2Fwmfg+K7K75vIw/f3Ytvgf6RznKeQ3L32DASSYA1eSIfq1Ik3tPL3g4LUXfaWNNLCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T09:36:48.042656Z"},"content_sha256":"8d78c8fc58da860cf524aecd4a2aa446fe577138821c40b68bf3836271829629","schema_version":"1.0","event_id":"sha256:8d78c8fc58da860cf524aecd4a2aa446fe577138821c40b68bf3836271829629"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:YLAUA24JES6YV253KVP5FLQ43N","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Play Favorites: A Statistical Method to Measure Self-Bias in LLM-as-a-Judge","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Evangelia Spiliopoulou, Graham Horwood, Hanna Burnsky, Jie Ma, Miguel Ballesteros, Riccardo Fogliato, Tamer Soliman","submitted_at":"2025-08-08T21:22:12Z","abstract_excerpt":"Large language models (LLMs) can serve as judges that offer rapid and reliable assessments of other LLM outputs. However, models may systematically assign overly favorable ratings to their own outputs, a phenomenon known as self-bias, which can distort evaluations of true model performance. Previous studies often conflate genuine differences in model quality with bias or incorrectly assume that evaluations from LLMs and humans follow the same rating distributions. In this work, we present a statistical framework that explicitly formalizes assumptions under which self-bias can be identified and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.06709","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/2508.06709/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":"2026-07-05T11:51:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qaG6DKZcD44/2chw8l8QANbjtGYNxXAI83kwDPSh2nU49czAqnmY8ywcxyt4XC+92KPUmRPANZvbvzpo/o5lAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T09:36:48.043219Z"},"content_sha256":"4b89a493b790ce4a211ca38520ac805873ba147e9fb0fedcf797f7c11167f504","schema_version":"1.0","event_id":"sha256:4b89a493b790ce4a211ca38520ac805873ba147e9fb0fedcf797f7c11167f504"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YLAUA24JES6YV253KVP5FLQ43N/bundle.json","state_url":"https://pith.science/pith/YLAUA24JES6YV253KVP5FLQ43N/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YLAUA24JES6YV253KVP5FLQ43N/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":"2026-08-18T09:36:48Z","links":{"resolver":"https://pith.science/pith/YLAUA24JES6YV253KVP5FLQ43N","bundle":"https://pith.science/pith/YLAUA24JES6YV253KVP5FLQ43N/bundle.json","state":"https://pith.science/pith/YLAUA24JES6YV253KVP5FLQ43N/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YLAUA24JES6YV253KVP5FLQ43N/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:YLAUA24JES6YV253KVP5FLQ43N","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":"f3b3979d516d168db47d86b63501621e7db8d5393025a0716ff81ebf7701a8f6","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-08T21:22:12Z","title_canon_sha256":"883df8c3e2ec65e16541d809357732cd712c0dc41c85ac87c67dae6b36e12bb6"},"schema_version":"1.0","source":{"id":"2508.06709","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.06709","created_at":"2026-07-05T11:51:05Z"},{"alias_kind":"arxiv_version","alias_value":"2508.06709v1","created_at":"2026-07-05T11:51:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.06709","created_at":"2026-07-05T11:51:05Z"},{"alias_kind":"pith_short_12","alias_value":"YLAUA24JES6Y","created_at":"2026-07-05T11:51:05Z"},{"alias_kind":"pith_short_16","alias_value":"YLAUA24JES6YV253","created_at":"2026-07-05T11:51:05Z"},{"alias_kind":"pith_short_8","alias_value":"YLAUA24J","created_at":"2026-07-05T11:51:05Z"}],"graph_snapshots":[{"event_id":"sha256:4b89a493b790ce4a211ca38520ac805873ba147e9fb0fedcf797f7c11167f504","target":"graph","created_at":"2026-07-05T11:51:05Z","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/2508.06709/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) can serve as judges that offer rapid and reliable assessments of other LLM outputs. However, models may systematically assign overly favorable ratings to their own outputs, a phenomenon known as self-bias, which can distort evaluations of true model performance. Previous studies often conflate genuine differences in model quality with bias or incorrectly assume that evaluations from LLMs and humans follow the same rating distributions. In this work, we present a statistical framework that explicitly formalizes assumptions under which self-bias can be identified and","authors_text":"Evangelia Spiliopoulou, Graham Horwood, Hanna Burnsky, Jie Ma, Miguel Ballesteros, Riccardo Fogliato, Tamer Soliman","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-08T21:22:12Z","title":"Play Favorites: A Statistical Method to Measure Self-Bias in LLM-as-a-Judge"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.06709","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":"sha256:8d78c8fc58da860cf524aecd4a2aa446fe577138821c40b68bf3836271829629","target":"record","created_at":"2026-07-05T11:51:05Z","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":"f3b3979d516d168db47d86b63501621e7db8d5393025a0716ff81ebf7701a8f6","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-08T21:22:12Z","title_canon_sha256":"883df8c3e2ec65e16541d809357732cd712c0dc41c85ac87c67dae6b36e12bb6"},"schema_version":"1.0","source":{"id":"2508.06709","kind":"arxiv","version":1}},"canonical_sha256":"c2c1406b8924bd8aebbb555fd2ae1cdb56c26663272e463e152f88fd441e261d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c2c1406b8924bd8aebbb555fd2ae1cdb56c26663272e463e152f88fd441e261d","first_computed_at":"2026-07-05T11:51:05.313415Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:51:05.313415Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"n9d+NOo7IxwzbhHeo3ZDXuIHcFo2xfRaks3gp7MW+z6a+Hq0GIpxbFokdo/0M53nKrvD72PA211Y3tDwFqzcBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:51:05.314043Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.06709","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8d78c8fc58da860cf524aecd4a2aa446fe577138821c40b68bf3836271829629","sha256:4b89a493b790ce4a211ca38520ac805873ba147e9fb0fedcf797f7c11167f504"],"state_sha256":"4c546485b007d19cb26390ad970ec07d532c6ba8e7988ea5513b1483b6c08fd9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DiHo5HPoEA7sLiJ0JxQcBftzCrIcjobtIAScH2+tt7TBtWuVj19tQleZkcBO4yWnWMfkSkmRBSDha8xm5zadAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T09:36:48.048525Z","bundle_sha256":"cb57d6e78226692a7147d22236be53b910c36bba39849e434bae0183380d9f50"}}