News Topic
LLM Evaluation & Leaderboards
News on LLM evaluation and benchmarks, covering new tests, methodology changes, leaderboard results, and whether findings can be reproduced.
Featured
Latest
NewsVals finds prompt framing shifts AI targeting in simulations
NewsPerplexity releases BrowseSafe++ benchmark for browser-agent prompt injection
NewsCWE-bench v2 opens 100 public security-repair tasks
NewsAllen Institute for AI releases StateBench for long-document state tracking
NewsAi2 releases XYBench to test how assistants handle mistaken premises
NewsApple study finds MMLU weakly tracks LLM judge quality
NewsMBZUAI introduces MedRoundsQA benchmark for diagnostic conversations
NewsVals finds DeepSeek trails GPT-6 Sol Daybreak by 8 points in simulated cyber test
NewsOpenAI introduces LifeVal benchmark for everyday AI tasks
NewsProperty Finder rebuilds AI benchmark, reports 86% lower inference cost
NewsIBM and Kaggle launch maintained ScarfBench Java migration leaderboard
NewsiMerit details Scholars workflows for specialized AI model evaluation


