9 papers
Adapting Vision Foundation Models with Cascaded Semantics
Xi Xiao, Xingjian Li, Cheng Han +8
Prompt tuning, a leading parameter-efficient adaptation paradigm in NLP, has recently been extended to computer vision. Visual prompt tuning (VPT) adapts pre-trained vision transfo…
Layer-Specific Prompt Fusion Discovery via Differentiable Search in Vision Foundation Models
Xi Xiao, Xingjian Li, Yunbei Zhang +7
Visual prompt tuning has emerged as a parameter-efficient fine-tuning approach for adapting large-scale Vision Transformers (ViTs) to downstream tasks. As its learnable prompts are…
SciTrace: Trajectory-Aware Safety Reasoning for Scientific Discovery Agents
Tanush Swaminathan, Runmin Jiang, Letian Zhang +1
LLM-based scientific agents have shown strong capacity for autonomous research, yet their safety layers remain structurally divorced from core reasoning: they inspect pipeline outp…
Accelerating Eigenvalue Dataset Generation via Chebyshev Subspace Filter
Hong Wang, Jie Wang, Jian Luo +4
Eigenvalue problems are among the most important topics in many scientific disciplines. With the recent surge and development of machine learning, neural eigenvalue methods have at…
Mind the Rarities: Can Rare Skin Diseases Be Reliably Diagnosed via Diagnostic Reasoning?
Yang Liu, Jiyao Yang, Hongjin Zhao +10
Large vision-language models (LVLMs) demonstrate strong performance in dermatology; however, evaluating diagnostic reasoning for rare conditions remains largely unexplored. Existin…
SG-OIF: A Stability-Guided Online Influence Framework for Reliable Vision Data
Penghao Rao, Runmin Jiang, Min Xu
Approximating training-point influence on test predictions is critical for deploying deep-learning vision models, essential for locating noisy data. Though the influence function w…