Showing cs.CLShow all
3 papers · 1 filter
cs.CL2026
BAS: A Decision-Theoretic Approach to Evaluating Large Language Model Confidence
Sean Wu, Fredrik K. Gustafsson, Edward Phillips +3
Large language models (LLMs) often produce confident but incorrect answers in settings where abstention would be safer. Standard evaluation protocols, however, require a response a…
cs.CL2026
Entropy Alone is Insufficient for Safe Selective Prediction in LLMs
Edward Phillips, Fredrik K. Gustafsson, Sean Wu +2
Selective prediction systems can mitigate harms resulting from language model hallucinations by abstaining from answering in high-risk cases. Uncertainty quantification techniques…
cs.CL2025
Geometric Uncertainty for Detecting and Correcting Hallucinations in LLMs
Edward Phillips, Sean Wu, Soheila Molaei +3
Large language models demonstrate impressive results across diverse tasks but are still known to hallucinate, generating linguistically plausible but incorrect answers to questions…