# Training an LLM-as-a-Judge Model: Pipeline, Insights, and Practical Lessons

arXiv: https://arxiv.org/abs/2502.02988 · HTML: https://mlsanity.com/paper/2502.02988/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 31 of its 56 references that are in the map.
- First posted on arXiv 5 Feb 2025.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)
- Knowledge distillation / [Knowledge distillation for LLMs](https://mlsanity.com/distill/knowledge-distillation-for-llms/)

## Builds on (most cited first, 30 of 31)

- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Self-Instruct: Aligning Language Models with Self-Generated Instructions](https://mlsanity.com/paper/2212.10560/) (2022)
- [ZeRO: Memory Optimizations Toward Training Trillion Parameter Models](https://mlsanity.com/paper/1910.02054/) (2019)
- [Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference](https://mlsanity.com/paper/2403.04132/) (2024)
- [AGIEval: A Human-Centric Benchmark for Evaluating Foundation Models](https://mlsanity.com/paper/2304.06364/) (2023)
- [From Crowdsourced Data to High-Quality Benchmarks: Arena-Hard and BenchBuilder Pipeline](https://mlsanity.com/paper/2406.11939/) (2024)
- [Orca: Progressive Learning from Complex Explanation Traces of GPT-4](https://mlsanity.com/paper/2306.02707/) (2023)
- [From Quantity to Quality: Boosting LLM Performance with Self-Guided Data Selection for Instruction Tuning](https://mlsanity.com/paper/2308.12032/) (2023)
- [TrustLLM: Trustworthiness in Large Language Models](https://mlsanity.com/paper/2401.05561/) (2024)
- [The Unlocking Spell on Base LLMs: Rethinking Alignment via In-Context Learning](https://mlsanity.com/paper/2312.01552/) (2023)
- [How Abilities in Large Language Models are Affected by Supervised Fine-tuning Data Composition](https://mlsanity.com/paper/2310.05492/) (2023)
- [WildBench: Benchmarking LLMs with Challenging Tasks from Real Users in the Wild](https://mlsanity.com/paper/2406.04770/) (2024)
- [Data Mixing Laws: Optimizing Data Mixtures by Predicting Language Modeling Performance](https://mlsanity.com/paper/2403.16952/) (2024)
- [Fine-Tuning or Retrieval? Comparing Knowledge Injection in LLMs](https://mlsanity.com/paper/2312.05934/) (2023)
- [A Study of Generative Large Language Model for Medical Research and Healthcare](https://mlsanity.com/paper/2305.13523/) (2023)
- [Cognitive Bias in Decision-Making with LLMs](https://mlsanity.com/paper/2403.00811/) (2024)
- [How Do Large Language Models Acquire Factual Knowledge During Pretraining?](https://mlsanity.com/paper/2406.11813/) (2024)
- [Foundational Autoraters: Taming Large Language Models for Better Automatic Evaluation](https://mlsanity.com/paper/2407.10817/) (2024)
- [CritiqueLLM: Towards an Informative Critique Generation Model for Evaluation of Large Language Model Generation](https://mlsanity.com/paper/2311.18702/) (2023)
- [SmallToLarge (S2L): Scalable Data Selection for Fine-tuning Large Language Models by Summarizing Training Trajectories of Small Models](https://mlsanity.com/paper/2403.07384/) (2024)
- [Assessing and Understanding Creativity in Large Language Models](https://mlsanity.com/paper/2401.12491/) (2024)
- [DyKnow: Dynamically Verifying Time-Sensitive Factual Knowledge in LLMs](https://mlsanity.com/paper/2404.08700/) (2024)
- [Balanced Data Sampling for Language Model Training with Clustering](https://mlsanity.com/paper/2402.14526/) (2024)
- [Elephants Never Forget: Testing Language Models for Memorization of Tabular Data](https://mlsanity.com/paper/2403.06644/) (2024)
- [Raising the Bar: Investigating the Values of Large Language Models via Generative Evolving Testing](https://mlsanity.com/paper/2406.14230/) (2024)
- [RefGPT: Dialogue Generation of GPT, by GPT, and for GPT](https://mlsanity.com/paper/2305.14994/) (2023)

## Built on by (most cited first, 9 of 9)

- [Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models](https://mlsanity.com/paper/2503.09567/) (2025)
- [Structural Reward Model: Enhancing Interpretability, Efficiency, and Scalability in Reward Modeling](https://mlsanity.com/paper/2509.25361/) (2025)
- [Advanced Tool Learning and Selection System (ATLASS): A Closed-Loop Framework Using LLM](https://mlsanity.com/paper/2503.10071/) (2025)
- [Reference-Free Rating of LLM Responses via Latent Information](https://mlsanity.com/paper/2509.24678/) (2025)
- [Large Language Models Could Be Rote Learners](https://mlsanity.com/paper/2504.08300/) (2025)
- [Noise or Nuance: An Investigation Into Useful Information and Filtering For LLM Driven AKBC](https://mlsanity.com/paper/2509.08903/) (2025)
- [CRACQ: A Multi-Dimensional Approach To Automated Document Assessment](https://mlsanity.com/paper/2510.02337/) (2025)
- [SAGE: Scalable AI Governance & Evaluation](https://mlsanity.com/paper/2602.07840/) (2026)
- [LLM-as-a-Discriminator: When Synthetic Tables Still Look Real](https://mlsanity.com/paper/2606.09865/) (2026)

## Abstract (arXiv)

The rapid advancement of large language models (LLMs) has opened new possibilities for their adoption as evaluative judges. This paper introduces Themis, a fine-tuned LLM judge that delivers sophisticated context-aware evaluations. We provide a comprehensive overview of the development pipeline for Themis, highlighting its scenario-dependent evaluation prompts and two novel methods for controlled instruction generation. These designs enable Themis to effectively distill evaluative skills from teacher models, while retaining flexibility for continuous development. We introduce two human-labeled benchmarks for meta-evaluation, demonstrating that Themis can achieve high alignment with human preferences in an economical manner. Additionally, we explore insights into the LLM-as-a-judge paradigm, revealing nuances in performance and the varied effects of reference answers. Notably, we observe that pure knowledge distillation from strong LLMs, though common, does not guarantee performance improvement through scaling. We propose a mitigation strategy based on instruction-following difficulty. Furthermore, we provide practical guidelines covering data balancing, prompt customization, multi-objective training, and metric aggregation. We aim for our method and findings, along with the fine-tuning data, benchmarks, and model checkpoints, to support future research and development in this area.