7 papers
K-Prism: A Knowledge-Guided and Prompt Integrated Universal Medical Image Segmentation Model
Bangwei Guo, Yunhe Gao, Meng Ye +4
Medical image segmentation is fundamental to clinical decision-making, yet existing models remain fragmented. They are usually trained on single knowledge sources and specific to i…
Evidence Over Plans: Online Trajectory Verification for Skill Distillation
Yang Zhou, Zihan Dong, Zhenting Wang +7
Agent skills can remarkably improve task success rates by using human-written procedural documents, but their quality is difficult to assess without environment-grounded verificati…
Can Cross-Layer Transcoders Replace Vision Transformer Activations? An Interpretable Perspective on Vision
Gerasimos Chatzoudis, Konstantinos D. Polyzos, Zhuowei Li +4
Understanding the internal activations of Vision Transformers (ViTs) is critical for building interpretable and trustworthy models. While Sparse Autoencoders (SAEs) have been used…
LUCID-SAE: Learning Unified Vision-Language Sparse Codes for Interpretable Concept Discovery
Difei Gu, Yunhe Gao, Gerasimos Chatzoudis +6
Sparse autoencoders (SAEs) offer a natural path toward comparable explanations across different representation spaces. However, current SAEs are trained per modality, producing dic…
M^3-Bench: Multi-Modal, Multi-Hop, Multi-Threaded Tool-Using MLLM Agent Benchmark
Yang Zhou, Mingyu Zhao, Zhenting Wang +6
We present M^3-Bench, the first benchmark for evaluating multimodal tool use under the Model Context Protocol. The benchmark targets realistic, multi-hop and multi-threaded workflo…
Anatomy-VLM: A Fine-grained Vision-Language Model for Medical Interpretation
Difei Gu, Yunhe Gao, Mu Zhou +1
Accurate disease interpretation from radiology remains challenging due to imaging heterogeneity. Achieving expert-level diagnostic decisions requires integration of subtle image fe…