5 papers
Confidence Should Be Calibrated More Than One Turn Deep
Zhaohan Zhang, Chengzhengxu Li, Xiaoming Liu +3
Large Language Models (LLMs) are increasingly applied in high-stakes domains such as finance, healthcare, and education, where reliable multi-turn interactions with users are essen…
Beyond Linear Probes: Dynamic Safety Monitoring for Language Models
James Oldfield, Philip Torr, Ioannis Patras +2
Monitoring large language models' (LLMs) activations is an effective way to detect harmful requests before they lead to unsafe outputs. However, traditional safety monitors often r…
GrACE: A Generative Approach to Better Confidence Elicitation and Efficient Test-Time Scaling in Large Language Models
Zhaohan Zhang, Ziquan Liu, Ioannis Patras
Assessing the reliability of Large Language Models (LLMs) by confidence elicitation is a prominent approach to AI safety in high-stakes applications, such as healthcare and finance…
Towards Interpretability Without Sacrifice: Faithful Dense Layer Decomposition with Mixture of Decoders
James Oldfield, Shawn Im, Sharon Li +3
Multilayer perceptrons (MLPs) are an integral part of large language models, yet their dense representations render them difficult to understand, edit, and steer. Recent methods le…
ReWind: Understanding Long Videos with Instructed Learnable Memory
Anxhelo Diko, Tinghuai Wang, Wassim Swaileh +2
Vision-Language Models (VLMs) are crucial for applications requiring integrated understanding textual and visual information. However, existing VLMs struggle with long videos due t…