{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:PVGXLFGYQPO4ABUOWT4XYTKXSN","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":"1e2ab7d11667ebcafd23e4c4df8171135c31701c2c36404db5fd95a8e6cf4ac4","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-03T18:48:38Z","title_canon_sha256":"6c465bcb02f8184bce92523e2ee6de6177efcf3d30bdd2b794b9ed219ef969fe"},"schema_version":"1.0","source":{"id":"2410.02902","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.02902","created_at":"2026-07-05T10:19:56Z"},{"alias_kind":"arxiv_version","alias_value":"2410.02902v4","created_at":"2026-07-05T10:19:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.02902","created_at":"2026-07-05T10:19:56Z"},{"alias_kind":"pith_short_12","alias_value":"PVGXLFGYQPO4","created_at":"2026-07-05T10:19:56Z"},{"alias_kind":"pith_short_16","alias_value":"PVGXLFGYQPO4ABUO","created_at":"2026-07-05T10:19:56Z"},{"alias_kind":"pith_short_8","alias_value":"PVGXLFGY","created_at":"2026-07-05T10:19:56Z"}],"graph_snapshots":[{"event_id":"sha256:a840c4df73dcd631543deb573fc05432b4aa35a5c395ef8890b8a59ccab14b26","target":"graph","created_at":"2026-07-05T10:19:56Z","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/2410.02902/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"General-purpose LLM judges capable of human-level evaluation provide not only a scalable and accurate way of evaluating instruction-following LLMs but also new avenues for supervising and improving their performance. One promising way of leveraging LLM judges for supervision is through Minimum Bayes Risk (MBR) decoding, which uses a reference-based evaluator to select a high-quality output from amongst a set of candidate outputs. In the first part of this work, we explore using MBR decoding as a method for improving the test-time performance of instruction-following LLMs. We find that MBR deco","authors_text":"Amanda Bertsch, Graham Neubig, Ian Wu, Patrick Fernandes, Seungone Kim, Sina Pakazad","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-03T18:48:38Z","title":"Better Instruction-Following Through Minimum Bayes Risk"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.02902","kind":"arxiv","version":4},"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":"sha256:6c2d1f18b187359ae1191ef6289261fbd1fcf03b9907705d10eebbb3740f3c03","target":"record","created_at":"2026-07-05T10:19:56Z","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":"1e2ab7d11667ebcafd23e4c4df8171135c31701c2c36404db5fd95a8e6cf4ac4","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-03T18:48:38Z","title_canon_sha256":"6c465bcb02f8184bce92523e2ee6de6177efcf3d30bdd2b794b9ed219ef969fe"},"schema_version":"1.0","source":{"id":"2410.02902","kind":"arxiv","version":4}},"canonical_sha256":"7d4d7594d883ddc0068eb4f97c4d579376f738e68037347c223b52a4b7d1557d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7d4d7594d883ddc0068eb4f97c4d579376f738e68037347c223b52a4b7d1557d","first_computed_at":"2026-07-05T10:19:56.007441Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:19:56.007441Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+IoNbsRw72D24+FANE6kVam/cvSJDoAUr1NC0Pja/mUHC5aJHr950IFG3iGVr3kgbu9ClOycbC42S258JeQxCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:19:56.007915Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.02902","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6c2d1f18b187359ae1191ef6289261fbd1fcf03b9907705d10eebbb3740f3c03","sha256:a840c4df73dcd631543deb573fc05432b4aa35a5c395ef8890b8a59ccab14b26"],"state_sha256":"ed67f1aea761d1dcb38fcc35c8987cfb58db49120d30d05d3e13bf86286910b1"}