mlsanityData to 7 Oct 2026

TRACT: Regression-Aware Fine-tuning Meets Chain-of-Thought Reasoning for LLM-as-a-Judge

arXiv 2503.04381 ↗First posted 6 Mar 2025cs.CLcited by 16

On the map: LLM reasoning → Reasoning benchmarks and evaluation

Abstract (arXiv)

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 with CoT), a method combining CoT reasoning with regression-aware training. TRACT consists of two stages: first, seed LLM is fine-tuned to generate CoTs, which serve as supervision for the second stage fine-tuning. The training objective of TRACT combines the CE loss for learning the CoT reasoning capabilities, and the regression-aware loss for the score prediction. Experiments across four LLM-as-a-judge datasets and two LLMs show that TRACT significantly outperforms existing methods. Extensive ablation studies validate the importance of each component in TRACT.

Key facts. Cited by 16 papers in the map (data through 7 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 15 of its 58 references that are in the map. First posted on arXiv 6 Mar 2025.

Builds on

15 of 58 references.

References · 15

  1. 2024The Llama 3 Herd of Modelscited by 14,192
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  3. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
  4. 2023Mistral 7Bcited by 3,065
  5. 2024Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phonecited by 1,883
  6. 2024LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Modelscited by 1,442
  7. 2024Tulu 3: Pushing Frontiers in Open Language Model Post-Trainingcited by 988
  8. 2024Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluatorscited by 865
  9. 2023Reinforced Self-Training (ReST) for Language Modelingcited by 486
  10. 2024Skywork-Reward: Bag of Tricks for Reward Modeling in LLMscited by 322
  11. 2024Critique-out-Loud Reward Modelscited by 105
  12. 2024Quantile Regression for Distributional Reward Models in RLHFcited by 59
3 more
  1. 2023Rewarding Chatbots for Real-World Engagement with Millions of Userscited by 58
  2. 2017Ensemble Distillation for Neural Machine Translationcited by 39
  3. 2021Self-Distillation Mixup Training for Non-autoregressive Neural Machine Translationcited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 8; 2026: 8.

16 papers.

Sort:

By area:

  1. 2025DeSTA2.5-Audio: Toward General-Purpose Large Audio Language Model with Self-Generated Cross-Modal Alignmentcited by 47
  2. 2025Evaluating Scoring Bias in LLM-as-a-Judgecited by 42
  3. 2026Diagnosing the Reliability of LLM-as-a-Judge via Item Response Theorycited by 13
  4. 2025A Survey of Scaling in Large Language Model Reasoningcited by 10
  5. 2025Quantitative LLM Judgescited by 6
  6. 2026DynSess: Dynamic Session-Level Evaluation and Optimization Framework for Role-Playing Agentscited by 2
  7. 2025Beyond Token-level Supervision: Unlocking the Potential of Decoding-based Regression via Reinforcement Learningcited by 1
  8. 2026REAL: Regression-Aware Reinforcement Learning for LLM-as-a-Judgecited by 1
  9. 2026RELISH: LLM REgression with a Latent Iterative State Headcited by 1
  10. 2026Auditing Stealth Sycophancy in Mental-Health Dialogue: Structured Clinical-State Diagnostics and Clean Matched Benchmarkscited by 1
  11. 2026Reinforcement Learning over Predictive Distributions for LLM Regressioncited by 1
  12. 2025Smart Trial: Evaluating the Use of Large Language Models for Recruiting Clinical Trial Participants via Social Mediacited by 0
4 more
  1. 2025Populism Meets AI: Advancing Populism Research with LLMscited by 0
  2. 2025Learning to Reason in LLMs by Expectation Maximizationcited by 0
  3. 2026Text-to-Distribution Prediction with Quantile Tokens and Neighbor Contextcited by 0
  4. 2026Biological Reasoning-Informed Regression for Interpretable Regulatory DNA Activity Predictioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025JudgeLRM: Large Reasoning Models as a Judge
  2. 2024From Calculation to Adjudication: Examining LLM judges on Mathematical Reasoning Tasks
  3. 2026JudgeStealer: Extracting LLM Judging Capabilities across Evaluation Protocols
  4. 2026Can LLM Reasoning Be Trusted? A Comparative Study: Using Human Benchmarking on Statistical Tasks
  5. 2025Improve LLM-as-a-Judge Ability as a General Ability
  6. 2025Rethinking Supervised Fine-Tuning: Emphasizing Key Answer Tokens for Improved LLM Accuracy
  7. 2023Scaling Relationship on Learning Mathematical Reasoning with Large Language Models
  8. 2026LLM-as-Judge on a Budget