5 papers
Detecting Misbehaviors of Large Vision-Language Models by Evidential Uncertainty Quantification
Tao Huang, Rui Wang, Xiaofei Liu +3
%Large vision-language models (LVLMs) have shown substantial advances in multimodal understanding and generation. However, when presented with incompetent or adversarial inputs, th…
State Rank Dynamics in Linear Attention LLMs
Ao Sun, Hongtao Zhang, Heng Zhou +9
Linear Attention Large Language Models (LLMs) offer a compelling recurrent formulation that compresses context into a fixed-size state matrix, enabling constant-time inference. How…
BMGQ: A Bottom-up Method for Generating Complex Multi-hop Reasoning Questions from Semi-structured Data
Bingsen Qiu, Zijian Liu, Xiao Liu +6
Building training-ready multi-hop question answering (QA) datasets that truly stress a model's retrieval and reasoning abilities remains highly challenging recently. While there ha…
SAM Encoder Breach by Adversarial Simplicial Complex Triggers Downstream Model Failures
Yi Qin, Rui Wang, Tao Huang +2
While the Segment Anything Model (SAM) transforms interactive segmentation with zero-shot abilities, its inherent vulnerabilities present a single-point risk, potentially leading t…
FoAM: Foresight-Augmented Multi-Task Imitation Policy for Robotic Manipulation
Litao Liu, Wentao Wang, Yifan Han +5
Multi-task imitation learning (MTIL) has shown significant potential in robotic manipulation by enabling agents to perform various tasks using a single policy. This simplifies the…