2 papers
cs.LG2025
Large Language Models Must Be Taught to Know What They Don't Know
Sanyam Kapoor, Nate Gruver, Manley Roberts +7
When using large language models (LLMs) in high-stakes applications, we need to know when we can trust their predictions. Some works argue that prompting high-performance LLMs is s…
cs.CL2025
LiveBench: A Challenging, Contamination-Limited LLM Benchmark
Colin White, Samuel Dooley, Manley Roberts +15
Test set contamination, wherein test data from a benchmark ends up in a newer model's training set, is a well-documented obstacle for fair LLM evaluation and can quickly render ben…