← Search

Jiawei Cao

9 accepted papers

2026

Hallucination Begins Where Saliency Drops

ICLR 2026oral

Recent studies have investigated attention dynamics in large vision language models (LVLMs), yet existing methods remain limited in reliably distinguishing hallucinated from correct outputs — primarily because they rely solely on forward-pass attention, ignoring gradient-based signals that reveal ho…

Cited by 0SourcecodeScholar
2026

Illuminating Visual Identity in Universal Multimodal Embeddings

CVPR 2026

Universal Multimodal Embeddings (UMEs) aim to unify various modalities and tasks into a shared representation space. In recent years, this field has witnessed substantial progress driven by the development of Multimodal Large Language Models (MLLMs). However, a crucial capability, visual identity di

Cited by 0SourcecodeScholar
2025

EFDTR: Learnable Elliptical Fourier Descriptor Transformer for Instance Segmentation

ICML 2025poster

Polygon-based object representations efficiently model object boundaries but are limited by high optimization complexity, which hinders their adoption compared to more flexible pixel-based methods. In this paper, we introduce a novel vertex regression loss grounded in Fourier elliptic descriptors,…

2025

Shallow Focus, Deep Fixes: Enhancing Shallow Layers Vision Attention Sinks to Alleviate Hallucination in LVLMs

EMNLP 2025

Multimodal large language models (MLLMs) demonstrate excellent abilities for understanding visual information, while the hallucination remains. Albeit image tokens constitute the majority of the MLLMs input, the relation between image tokens and hallucinations is still unexplored. In this paper, we

Cited by 0SourcePDFScholar
2019

Deep Reinforcement Learning in Soft Viscoelastic Actuator of Dielectric Elastomer

RA-L 2019

Dielectric elastomer actuators (DEAs) have been widely employed as artificial muscles in soft robots. Due to material viscoelasticity and nonlinear electromechanical coupling, it is challenging to accurately model a viscoelastic DEA, especially when the actuator is of a complex or irregular configur

Cited by 32SourceScholar
2018

Modelling and Control of a Novel Soft Crawling Robot Based on a Dielectric Elastomer Actuator

ICRA 2018poster

Soft robots have recently evoked extensive attention due to their abilities to work effectively in unstructured environments. As an actuation technology of soft robots, dielectric elastomers exhibit many intriguing attributes such as large strain and high energy density. This work presents a novel d…

Cited by 34SourceScholar
2018

Topology Optimized Design, Fabrication, and Characterization of a Soft Cable-Driven Gripper

RA-L 2018

Soft-bodied robots, due to their intrinsic compliance, have shown great potential for operating within unstructured environment and interacting with unknown objects. This letter deals with automatic design and fabrication of soft robots. From a structure point of view, we synthesize a soft cable-dri

Cited by 125SourceScholar
2017

Force Measurement Toward the Instability Theory of Soft Pneumatic Actuators

RA-L 2017

Silicone-based bending soft pneumatic actuators (SPAs) have been very popular, since they provide solutions to many applications that require comfort and safety. However, their further utilization seems to be thwarted due to their limited force output. Force output can be the most important property

Cited by 36SourceScholar
2017

Networked soft actuators with large deformations

ICRA 2017poster

Soft actuators play an important role in producing motions in soft robots, and dielectric elastomers have shown great promise because of their considerable voltage-induced deformation. In particular, air-filled dielectric elastomer actuators have been well studied, where the air inside provides pres…

Cited by 12SourceScholar