{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:BOFNM75YPDHYJTA5SDQZKWEXTC","short_pith_number":"pith:BOFNM75Y","canonical_record":{"source":{"id":"2503.04381","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-06T12:33:20Z","cross_cats_sorted":[],"title_canon_sha256":"d1293ef5b1dd7043c7c3041c004e77c1150ee545b49918b0c5555c6d94928528","abstract_canon_sha256":"3d8848b9fef45082c85eb3463a51d56b01467d83deb3cf84ad801a8ec8566409"},"schema_version":"1.0"},"canonical_sha256":"0b8ad67fb878cf84cc1d90e1955897989999399bf31b9fd8107810dd291b94a8","source":{"kind":"arxiv","id":"2503.04381","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.04381","created_at":"2026-07-05T11:16:56Z"},{"alias_kind":"arxiv_version","alias_value":"2503.04381v2","created_at":"2026-07-05T11:16:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.04381","created_at":"2026-07-05T11:16:56Z"},{"alias_kind":"pith_short_12","alias_value":"BOFNM75YPDHY","created_at":"2026-07-05T11:16:56Z"},{"alias_kind":"pith_short_16","alias_value":"BOFNM75YPDHYJTA5","created_at":"2026-07-05T11:16:56Z"},{"alias_kind":"pith_short_8","alias_value":"BOFNM75Y","created_at":"2026-07-05T11:16:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:BOFNM75YPDHYJTA5SDQZKWEXTC","target":"record","payload":{"canonical_record":{"source":{"id":"2503.04381","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-06T12:33:20Z","cross_cats_sorted":[],"title_canon_sha256":"d1293ef5b1dd7043c7c3041c004e77c1150ee545b49918b0c5555c6d94928528","abstract_canon_sha256":"3d8848b9fef45082c85eb3463a51d56b01467d83deb3cf84ad801a8ec8566409"},"schema_version":"1.0"},"canonical_sha256":"0b8ad67fb878cf84cc1d90e1955897989999399bf31b9fd8107810dd291b94a8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:16:56.393141Z","signature_b64":"G+M1XtszGn9ys3e7pIBVhDLp5dWhZ6pNvYPbnNSuPITGVgXVcCkKZI2UODfSClX4rWu7/WdGg68iLYpkXLhyCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0b8ad67fb878cf84cc1d90e1955897989999399bf31b9fd8107810dd291b94a8","last_reissued_at":"2026-07-05T11:16:56.392646Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:16:56.392646Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.04381","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:16:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qyeB3gzldE3mmso9VbF+tFrNVi9XrVoQmLX38ZB8N2V/j1xsYo4m0FALT+qh545SNutxvZvj+bYkEK8xN4ntAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T22:43:46.560743Z"},"content_sha256":"aaa90002e03d59810304fc6310756d3259844ac3c20721b63c6cc8eb3eaa0a7c","schema_version":"1.0","event_id":"sha256:aaa90002e03d59810304fc6310756d3259844ac3c20721b63c6cc8eb3eaa0a7c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:BOFNM75YPDHYJTA5SDQZKWEXTC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"TRACT: Regression-Aware Fine-tuning Meets Chain-of-Thought Reasoning for LLM-as-a-Judge","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cheng-Han Chiang, Hung-yi Lee, Michal Lukasik","submitted_at":"2025-03-06T12:33:20Z","abstract_excerpt":"The LLM-as-a-judge paradigm uses large language models (LLMs) for automated text evaluation, where a numerical assessment is assigned by an LLM to the input text following scoring rubrics. Existing methods for LLM-as-a-judge use cross-entropy (CE) loss for fine-tuning, which neglects the numeric nature of score prediction. Recent work addresses numerical prediction limitations of LLM fine-tuning through regression-aware fine-tuning, which, however, does not consider chain-of-thought (CoT) reasoning for score prediction. In this paper, we introduce TRACT (Two-stage Regression-Aware fine-tuning "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.04381","kind":"arxiv","version":2},"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/2503.04381/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":"2026-07-05T11:16:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ITkq1Z5vxxaK+9PpvXoxaNPJ8wjE7QH//N035oekgcvFmgRO8ieUGv4Be8fUWXsW94LYZ63UHEJkzQSNv4ctCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T22:43:46.561298Z"},"content_sha256":"4dbe1afb2a7f2f4fc493f299f76841e8c03c8388f2547056b79894668e2cb2d6","schema_version":"1.0","event_id":"sha256:4dbe1afb2a7f2f4fc493f299f76841e8c03c8388f2547056b79894668e2cb2d6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BOFNM75YPDHYJTA5SDQZKWEXTC/bundle.json","state_url":"https://pith.science/pith/BOFNM75YPDHYJTA5SDQZKWEXTC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BOFNM75YPDHYJTA5SDQZKWEXTC/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":"2026-08-07T22:43:46Z","links":{"resolver":"https://pith.science/pith/BOFNM75YPDHYJTA5SDQZKWEXTC","bundle":"https://pith.science/pith/BOFNM75YPDHYJTA5SDQZKWEXTC/bundle.json","state":"https://pith.science/pith/BOFNM75YPDHYJTA5SDQZKWEXTC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BOFNM75YPDHYJTA5SDQZKWEXTC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:BOFNM75YPDHYJTA5SDQZKWEXTC","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":"3d8848b9fef45082c85eb3463a51d56b01467d83deb3cf84ad801a8ec8566409","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-06T12:33:20Z","title_canon_sha256":"d1293ef5b1dd7043c7c3041c004e77c1150ee545b49918b0c5555c6d94928528"},"schema_version":"1.0","source":{"id":"2503.04381","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.04381","created_at":"2026-07-05T11:16:56Z"},{"alias_kind":"arxiv_version","alias_value":"2503.04381v2","created_at":"2026-07-05T11:16:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.04381","created_at":"2026-07-05T11:16:56Z"},{"alias_kind":"pith_short_12","alias_value":"BOFNM75YPDHY","created_at":"2026-07-05T11:16:56Z"},{"alias_kind":"pith_short_16","alias_value":"BOFNM75YPDHYJTA5","created_at":"2026-07-05T11:16:56Z"},{"alias_kind":"pith_short_8","alias_value":"BOFNM75Y","created_at":"2026-07-05T11:16:56Z"}],"graph_snapshots":[{"event_id":"sha256:4dbe1afb2a7f2f4fc493f299f76841e8c03c8388f2547056b79894668e2cb2d6","target":"graph","created_at":"2026-07-05T11:16:56Z","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/2503.04381/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The LLM-as-a-judge paradigm uses large language models (LLMs) for automated text evaluation, where a numerical assessment is assigned by an LLM to the input text following scoring rubrics. Existing methods for LLM-as-a-judge use cross-entropy (CE) loss for fine-tuning, which neglects the numeric nature of score prediction. Recent work addresses numerical prediction limitations of LLM fine-tuning through regression-aware fine-tuning, which, however, does not consider chain-of-thought (CoT) reasoning for score prediction. In this paper, we introduce TRACT (Two-stage Regression-Aware fine-tuning ","authors_text":"Cheng-Han Chiang, Hung-yi Lee, Michal Lukasik","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-06T12:33:20Z","title":"TRACT: Regression-Aware Fine-tuning Meets Chain-of-Thought Reasoning for LLM-as-a-Judge"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.04381","kind":"arxiv","version":2},"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:aaa90002e03d59810304fc6310756d3259844ac3c20721b63c6cc8eb3eaa0a7c","target":"record","created_at":"2026-07-05T11:16:56Z","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":"3d8848b9fef45082c85eb3463a51d56b01467d83deb3cf84ad801a8ec8566409","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-06T12:33:20Z","title_canon_sha256":"d1293ef5b1dd7043c7c3041c004e77c1150ee545b49918b0c5555c6d94928528"},"schema_version":"1.0","source":{"id":"2503.04381","kind":"arxiv","version":2}},"canonical_sha256":"0b8ad67fb878cf84cc1d90e1955897989999399bf31b9fd8107810dd291b94a8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0b8ad67fb878cf84cc1d90e1955897989999399bf31b9fd8107810dd291b94a8","first_computed_at":"2026-07-05T11:16:56.392646Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:16:56.392646Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"G+M1XtszGn9ys3e7pIBVhDLp5dWhZ6pNvYPbnNSuPITGVgXVcCkKZI2UODfSClX4rWu7/WdGg68iLYpkXLhyCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:16:56.393141Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.04381","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:aaa90002e03d59810304fc6310756d3259844ac3c20721b63c6cc8eb3eaa0a7c","sha256:4dbe1afb2a7f2f4fc493f299f76841e8c03c8388f2547056b79894668e2cb2d6"],"state_sha256":"b4cafd3c795f39974acce787e7b1fbb114f54f7d21c3595a9139f5ecd8a991a3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JtOZ4xLFXMSD4jNJWJL40RULYfka3ABhYEE1oALjsKO6jp8swG7i1oL3gMku6XPSSlO/7zNLMUTkG3srkD+1BQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T22:43:46.564960Z","bundle_sha256":"f333760828369780e6e3116025ccd1feaeb777081bdc4ac0dcce6da52f6a29b6"}}