{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:4TR6JJARK2RBIOZOHSNT7VBIZ2","short_pith_number":"pith:4TR6JJAR","canonical_record":{"source":{"id":"2309.13308","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-23T08:46:11Z","cross_cats_sorted":[],"title_canon_sha256":"1e41e3ea3459fcc779208f7888612d6b1c8f637677689fa60ce4b2b10f1c6ce1","abstract_canon_sha256":"33d76f133bfbf492b2da063b035a1e84a5d33e8eb2b2c816cb37ca8cf18747c5"},"schema_version":"1.0"},"canonical_sha256":"e4e3e4a41156a2143b2e3c9b3fd428cea3c0e5ff4983b14878e936988085db7b","source":{"kind":"arxiv","id":"2309.13308","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.13308","created_at":"2026-07-05T06:53:50Z"},{"alias_kind":"arxiv_version","alias_value":"2309.13308v1","created_at":"2026-07-05T06:53:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.13308","created_at":"2026-07-05T06:53:50Z"},{"alias_kind":"pith_short_12","alias_value":"4TR6JJARK2RB","created_at":"2026-07-05T06:53:50Z"},{"alias_kind":"pith_short_16","alias_value":"4TR6JJARK2RBIOZO","created_at":"2026-07-05T06:53:50Z"},{"alias_kind":"pith_short_8","alias_value":"4TR6JJAR","created_at":"2026-07-05T06:53:50Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:4TR6JJARK2RBIOZOHSNT7VBIZ2","target":"record","payload":{"canonical_record":{"source":{"id":"2309.13308","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-23T08:46:11Z","cross_cats_sorted":[],"title_canon_sha256":"1e41e3ea3459fcc779208f7888612d6b1c8f637677689fa60ce4b2b10f1c6ce1","abstract_canon_sha256":"33d76f133bfbf492b2da063b035a1e84a5d33e8eb2b2c816cb37ca8cf18747c5"},"schema_version":"1.0"},"canonical_sha256":"e4e3e4a41156a2143b2e3c9b3fd428cea3c0e5ff4983b14878e936988085db7b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:53:50.084156Z","signature_b64":"2sm+NP3iv/B4TJy7u7B9TUXukpXIYxE0KpepIKMRI8OuF+37Ld4qTEIE0zAGnsiZhQ8gedgM4sKbQtPxuWB+Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4e3e4a41156a2143b2e3c9b3fd428cea3c0e5ff4983b14878e936988085db7b","last_reissued_at":"2026-07-05T06:53:50.083645Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:53:50.083645Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2309.13308","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":"2026-07-05T06:53:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"N9e8Q2Or0yxrbBo5I10oT50bAV4ejl3ED96qlgUiWtqtXRej9uBH62mm6wpKePmF6z3Pec3O0NgXvFWhTkUdDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T06:17:19.240215Z"},"content_sha256":"bfdd54cf5ab64f8cfa1ebd08ebcad0a350f5531ff7b59ddaad6f71069314dc65","schema_version":"1.0","event_id":"sha256:bfdd54cf5ab64f8cfa1ebd08ebcad0a350f5531ff7b59ddaad6f71069314dc65"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:4TR6JJARK2RBIOZOHSNT7VBIZ2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Calibrating LLM-Based Evaluator","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Feng Sun, Furu Wei, Haizhen Huang, Qi Zhang, Shaohan Huang, Tianchi Yang, Weiwei Deng, Yuxuan Liu, Zihan Zhang","submitted_at":"2023-09-23T08:46:11Z","abstract_excerpt":"Recent advancements in large language models (LLMs) on language modeling and emergent capabilities make them a promising reference-free evaluator of natural language generation quality, and a competent alternative to human evaluation. However, hindered by the closed-source or high computational demand to host and tune, there is a lack of practice to further calibrate an off-the-shelf LLM-based evaluator towards better human alignment. In this work, we propose AutoCalibrate, a multi-stage, gradient-free approach to automatically calibrate and align an LLM-based evaluator toward human preference"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.13308","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/2309.13308/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":"2026-07-05T06:53:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QBgdvE+5Y0pb1l+48Evmlx9X+3NNyM9H5iZSw+n2G7caAr9165sg387xxb7+g3a+9q4/wFWUN/sYaQYo/GBECw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T06:17:19.241100Z"},"content_sha256":"fabbfcb999097a9b23a5794a27b2c7bf42b2d8340af11b6cb957a449df0214cc","schema_version":"1.0","event_id":"sha256:fabbfcb999097a9b23a5794a27b2c7bf42b2d8340af11b6cb957a449df0214cc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4TR6JJARK2RBIOZOHSNT7VBIZ2/bundle.json","state_url":"https://pith.science/pith/4TR6JJARK2RBIOZOHSNT7VBIZ2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4TR6JJARK2RBIOZOHSNT7VBIZ2/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":"2026-08-13T06:17:19Z","links":{"resolver":"https://pith.science/pith/4TR6JJARK2RBIOZOHSNT7VBIZ2","bundle":"https://pith.science/pith/4TR6JJARK2RBIOZOHSNT7VBIZ2/bundle.json","state":"https://pith.science/pith/4TR6JJARK2RBIOZOHSNT7VBIZ2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4TR6JJARK2RBIOZOHSNT7VBIZ2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:4TR6JJARK2RBIOZOHSNT7VBIZ2","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":"33d76f133bfbf492b2da063b035a1e84a5d33e8eb2b2c816cb37ca8cf18747c5","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-23T08:46:11Z","title_canon_sha256":"1e41e3ea3459fcc779208f7888612d6b1c8f637677689fa60ce4b2b10f1c6ce1"},"schema_version":"1.0","source":{"id":"2309.13308","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.13308","created_at":"2026-07-05T06:53:50Z"},{"alias_kind":"arxiv_version","alias_value":"2309.13308v1","created_at":"2026-07-05T06:53:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.13308","created_at":"2026-07-05T06:53:50Z"},{"alias_kind":"pith_short_12","alias_value":"4TR6JJARK2RB","created_at":"2026-07-05T06:53:50Z"},{"alias_kind":"pith_short_16","alias_value":"4TR6JJARK2RBIOZO","created_at":"2026-07-05T06:53:50Z"},{"alias_kind":"pith_short_8","alias_value":"4TR6JJAR","created_at":"2026-07-05T06:53:50Z"}],"graph_snapshots":[{"event_id":"sha256:fabbfcb999097a9b23a5794a27b2c7bf42b2d8340af11b6cb957a449df0214cc","target":"graph","created_at":"2026-07-05T06:53:50Z","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/2309.13308/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advancements in large language models (LLMs) on language modeling and emergent capabilities make them a promising reference-free evaluator of natural language generation quality, and a competent alternative to human evaluation. However, hindered by the closed-source or high computational demand to host and tune, there is a lack of practice to further calibrate an off-the-shelf LLM-based evaluator towards better human alignment. In this work, we propose AutoCalibrate, a multi-stage, gradient-free approach to automatically calibrate and align an LLM-based evaluator toward human preference","authors_text":"Feng Sun, Furu Wei, Haizhen Huang, Qi Zhang, Shaohan Huang, Tianchi Yang, Weiwei Deng, Yuxuan Liu, Zihan Zhang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-23T08:46:11Z","title":"Calibrating LLM-Based Evaluator"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.13308","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:bfdd54cf5ab64f8cfa1ebd08ebcad0a350f5531ff7b59ddaad6f71069314dc65","target":"record","created_at":"2026-07-05T06:53:50Z","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":"33d76f133bfbf492b2da063b035a1e84a5d33e8eb2b2c816cb37ca8cf18747c5","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-23T08:46:11Z","title_canon_sha256":"1e41e3ea3459fcc779208f7888612d6b1c8f637677689fa60ce4b2b10f1c6ce1"},"schema_version":"1.0","source":{"id":"2309.13308","kind":"arxiv","version":1}},"canonical_sha256":"e4e3e4a41156a2143b2e3c9b3fd428cea3c0e5ff4983b14878e936988085db7b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e4e3e4a41156a2143b2e3c9b3fd428cea3c0e5ff4983b14878e936988085db7b","first_computed_at":"2026-07-05T06:53:50.083645Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:53:50.083645Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2sm+NP3iv/B4TJy7u7B9TUXukpXIYxE0KpepIKMRI8OuF+37Ld4qTEIE0zAGnsiZhQ8gedgM4sKbQtPxuWB+Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T06:53:50.084156Z","signed_message":"canonical_sha256_bytes"},"source_id":"2309.13308","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bfdd54cf5ab64f8cfa1ebd08ebcad0a350f5531ff7b59ddaad6f71069314dc65","sha256:fabbfcb999097a9b23a5794a27b2c7bf42b2d8340af11b6cb957a449df0214cc"],"state_sha256":"d8719c9b09b506acd5ef68228a0ff6a9d6945f563c7c54c6cc66fdeac9c9b023"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kObUzYB8R9dgZkQGazgkVw4FIRJAQzaToIURCAV8J8zRyFys4qK9555glbGWn6Lvr2KR14ZyakM4ZYFJjYV0CA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-13T06:17:19.249471Z","bundle_sha256":"0d7c48b881b022bc820ccec5bb216906e6f17b1b3cdd6e500298a7501311627e"}}