This article discusses how fine-tuning open-source LLM judges using Direct Preference Optimization (DPO) can lead to performance that matches or exceeds GPT-5.2 in evaluating model outputs. The authors trained models like GPT-OSS 120B and Qwen 3 235B on human preference data, achieving better accuracy and efficiency at a lower cost.

LLM-SRBench is a new benchmark aimed at enhancing scientific equation discovery using large language models, featuring comprehensive evaluation methods and open-source implementation. It includes a structured setup guide for running and contributing new search methods, as well as the necessary configurations for various datasets. The benchmark has been recognized for its significance, being selected for oral presentation at ICML 2025.