5 papers
Contrastive Conceptor Activation Steering (COAST): Unlocking Vision-Language-Action Models through Hidden States
Miranda Muqing Miao, Subin Kim, Brandon Yang +1
Vision-Language-Action (VLA) models leverage powerful perceptual priors from web-scale Vision-Language Model (VLM) pre-training, yet they remain surprisingly brittle in practice, f…
Conceptors for Semantic Steering
Ilias Triantafyllopoulos, Young-Min Cho, Ren Tao +6
Activation-based steering provides control of LLM behavior at inference time, but the dominant paradigm reduces each concept to a single direction whose geometry is left largely un…
Closing the Confidence-Faithfulness Gap in Large Language Models
Miranda Muqing Miao, Lyle Ungar
Large language models (LLMs) tend to verbalize confidence scores that are largely detached from their actual accuracy, yet the geometric relationship governing this behavior remain…
Correctness-Optimized Residual Activation Lens (CORAL): Transferrable and Calibration-Aware Inference-Time Steering
Miranda Muqing Miao, Young-Min Cho, Lyle Ungar
Large language models (LLMs) exhibit persistent miscalibration, especially after instruction tuning and preference alignment. Modified training objectives can improve calibration,…
The Impact of Language Mixing on Bilingual LLM Reasoning
Yihao Li, Jiayi Xin, Miranda Muqing Miao +2
Proficient multilingual speakers often intentionally switch languages in the middle of a conversation. Similarly, recent reasoning-focused bilingual large language models (LLMs) wi…