Anastasios Nikolas Angelopoulos

EditBench: Evaluating LLM Abilities to Perform Real-World Instructed Code Edits

Search Arena: Analyzing Search-Augmented LLMs

Cost-Optimal Active AI Model Evaluation

Exploring and Mitigating Adversarial Manipulation of Voting-Based Leaderboards

How to Evaluate Reward Models for RLHF

Prompt-to-Leaderboard: Prompt-Adaptive LLM Evaluations

Copilot Arena: A Platform for Code LLM Evaluation in the Wild

AutoEval Done Right: Using Synthetic Data for Model Evaluation