{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:X6PJI6EDP6VSP762BDTLKKZ52U","short_pith_number":"pith:X6PJI6ED","canonical_record":{"source":{"id":"2606.08092","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-06T10:36:24Z","cross_cats_sorted":[],"title_canon_sha256":"cdf9a08720e4e566bdb4b747e288a85f0a9dd8631c2fa015b54c4168860efa83","abstract_canon_sha256":"7dc2dbb2c916f8f32f5079712d557b9936d660d2ccbb118bdfebf26f333cd5e3"},"schema_version":"1.0"},"canonical_sha256":"bf9e9478837fab27ffda08e6b52b3dd51530f6940261fd422cf1ce9707755ac9","source":{"kind":"arxiv","id":"2606.08092","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.08092","created_at":"2026-06-09T01:05:26Z"},{"alias_kind":"arxiv_version","alias_value":"2606.08092v1","created_at":"2026-06-09T01:05:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.08092","created_at":"2026-06-09T01:05:26Z"},{"alias_kind":"pith_short_12","alias_value":"X6PJI6EDP6VS","created_at":"2026-06-09T01:05:26Z"},{"alias_kind":"pith_short_16","alias_value":"X6PJI6EDP6VSP762","created_at":"2026-06-09T01:05:26Z"},{"alias_kind":"pith_short_8","alias_value":"X6PJI6ED","created_at":"2026-06-09T01:05:26Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:X6PJI6EDP6VSP762BDTLKKZ52U","target":"record","payload":{"canonical_record":{"source":{"id":"2606.08092","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-06T10:36:24Z","cross_cats_sorted":[],"title_canon_sha256":"cdf9a08720e4e566bdb4b747e288a85f0a9dd8631c2fa015b54c4168860efa83","abstract_canon_sha256":"7dc2dbb2c916f8f32f5079712d557b9936d660d2ccbb118bdfebf26f333cd5e3"},"schema_version":"1.0"},"canonical_sha256":"bf9e9478837fab27ffda08e6b52b3dd51530f6940261fd422cf1ce9707755ac9","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T01:05:26.059273Z","signature_b64":"DfPRQkgLiA58Ra0cwQuOy7L4bxzkfAtZ+DnwOjMQ0fjAq1NlB3uG4fwX0q149Km6CCiDzmerUAiOdqJCw94NAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bf9e9478837fab27ffda08e6b52b3dd51530f6940261fd422cf1ce9707755ac9","last_reissued_at":"2026-06-09T01:05:26.058834Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T01:05:26.058834Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.08092","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":"2026-06-09T01:05:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"D+7+jiGZ60ubEKLDc4WtDnF5pCl5m8bsUoRDWrZVhY6PZBt33oSOjEtLI/JAu5+KEZzFKL7HH+ImgGvBA/YpDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-28T14:31:51.672974Z"},"content_sha256":"c8070f675828be6ed9f8766b99f867a10a45339dbec0fe437f59956165c87bed","schema_version":"1.0","event_id":"sha256:c8070f675828be6ed9f8766b99f867a10a45339dbec0fe437f59956165c87bed"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:X6PJI6EDP6VSP762BDTLKKZ52U","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"When Languages Disagree: Self-Evolving Multilingual LLM Judges","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Wei Lu, Xiyan Fu","submitted_at":"2026-06-06T10:36:24Z","abstract_excerpt":"Multilingual LLM-as-a-judge is widely used to evaluate model outputs across languages, but suffers from cross-lingual inconsistency (Fu and Liu, 2025). Existing methods typically treat this inconsistency as noise and mitigate it through voting or aggregation. In this work, we instead show that multilingual inconsistency can provide complementary evaluation signals. Our oracle analysis finds that sampling judgments across languages yields a higher performance upper bound than single-language judging, indicating that different languages potentially include complementary judgments. Motivated by t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.08092","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/2606.08092/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":"2026-06-09T01:05:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vQ+ZADE9u52EaqMlqjf0ktboK6zUmj53ciVtVvPCP2h6RcVIdVjU0GvHRk0DjNGBFjQWlS9S/kU94Qa/OPnxAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-28T14:31:51.673362Z"},"content_sha256":"ecf44a48600b5891c6997a3d58631983bd5c25c014fd885a1dfef6c0ad108b10","schema_version":"1.0","event_id":"sha256:ecf44a48600b5891c6997a3d58631983bd5c25c014fd885a1dfef6c0ad108b10"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/X6PJI6EDP6VSP762BDTLKKZ52U/bundle.json","state_url":"https://pith.science/pith/X6PJI6EDP6VSP762BDTLKKZ52U/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/X6PJI6EDP6VSP762BDTLKKZ52U/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":"2026-06-28T14:31:51Z","links":{"resolver":"https://pith.science/pith/X6PJI6EDP6VSP762BDTLKKZ52U","bundle":"https://pith.science/pith/X6PJI6EDP6VSP762BDTLKKZ52U/bundle.json","state":"https://pith.science/pith/X6PJI6EDP6VSP762BDTLKKZ52U/state.json","well_known_bundle":"https://pith.science/.well-known/pith/X6PJI6EDP6VSP762BDTLKKZ52U/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:X6PJI6EDP6VSP762BDTLKKZ52U","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":"7dc2dbb2c916f8f32f5079712d557b9936d660d2ccbb118bdfebf26f333cd5e3","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-06T10:36:24Z","title_canon_sha256":"cdf9a08720e4e566bdb4b747e288a85f0a9dd8631c2fa015b54c4168860efa83"},"schema_version":"1.0","source":{"id":"2606.08092","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.08092","created_at":"2026-06-09T01:05:26Z"},{"alias_kind":"arxiv_version","alias_value":"2606.08092v1","created_at":"2026-06-09T01:05:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.08092","created_at":"2026-06-09T01:05:26Z"},{"alias_kind":"pith_short_12","alias_value":"X6PJI6EDP6VS","created_at":"2026-06-09T01:05:26Z"},{"alias_kind":"pith_short_16","alias_value":"X6PJI6EDP6VSP762","created_at":"2026-06-09T01:05:26Z"},{"alias_kind":"pith_short_8","alias_value":"X6PJI6ED","created_at":"2026-06-09T01:05:26Z"}],"graph_snapshots":[{"event_id":"sha256:ecf44a48600b5891c6997a3d58631983bd5c25c014fd885a1dfef6c0ad108b10","target":"graph","created_at":"2026-06-09T01:05:26Z","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/2606.08092/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multilingual LLM-as-a-judge is widely used to evaluate model outputs across languages, but suffers from cross-lingual inconsistency (Fu and Liu, 2025). Existing methods typically treat this inconsistency as noise and mitigate it through voting or aggregation. In this work, we instead show that multilingual inconsistency can provide complementary evaluation signals. Our oracle analysis finds that sampling judgments across languages yields a higher performance upper bound than single-language judging, indicating that different languages potentially include complementary judgments. Motivated by t","authors_text":"Wei Lu, Xiyan Fu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-06T10:36:24Z","title":"When Languages Disagree: Self-Evolving Multilingual LLM Judges"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.08092","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:c8070f675828be6ed9f8766b99f867a10a45339dbec0fe437f59956165c87bed","target":"record","created_at":"2026-06-09T01:05:26Z","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":"7dc2dbb2c916f8f32f5079712d557b9936d660d2ccbb118bdfebf26f333cd5e3","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-06T10:36:24Z","title_canon_sha256":"cdf9a08720e4e566bdb4b747e288a85f0a9dd8631c2fa015b54c4168860efa83"},"schema_version":"1.0","source":{"id":"2606.08092","kind":"arxiv","version":1}},"canonical_sha256":"bf9e9478837fab27ffda08e6b52b3dd51530f6940261fd422cf1ce9707755ac9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bf9e9478837fab27ffda08e6b52b3dd51530f6940261fd422cf1ce9707755ac9","first_computed_at":"2026-06-09T01:05:26.058834Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-09T01:05:26.058834Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"DfPRQkgLiA58Ra0cwQuOy7L4bxzkfAtZ+DnwOjMQ0fjAq1NlB3uG4fwX0q149Km6CCiDzmerUAiOdqJCw94NAQ==","signature_status":"signed_v1","signed_at":"2026-06-09T01:05:26.059273Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.08092","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c8070f675828be6ed9f8766b99f867a10a45339dbec0fe437f59956165c87bed","sha256:ecf44a48600b5891c6997a3d58631983bd5c25c014fd885a1dfef6c0ad108b10"],"state_sha256":"32b9bf524c05277dc28274e2eb218850783dd0603d73f24cbb36b7ef87bf8325"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8KzEkqheeAKOucKs2V6YJAq4L6/F2FSaoC6JM9FNe2Ddmz/xPrld1doxv9+b2fsOdq+FadmvOxvA//BwrqYHAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-28T14:31:51.675473Z","bundle_sha256":"a81d1913ffe40a7809be2d22c72b087beee52e88fd047b989130eb8aa5166d29"}}