6 papers
Trace2Policy: From Expert Behavior Traces to Self-Evolving Decision Agents
Junli Zha, Jinbo Wang, Chao Zhou +1
Decision rules that enterprise experts apply tacitly -- in auditing, compliance, and contract review -- can be systematically recovered and improved through iterative error analysi…
PrismFlow: Residual Dynamics for Flow Matching in Time-Series Generation
Junru Zhang, Lang Feng, Jinbo Wang +6
Generating high-quality time-series data is challenging because real-world signals often exhibit multimodal patterns and multiscale dynamics, including oscillations and high-freque…
Illusion-Aware Visual Preprocessing and Anti-Illusion Prompting for Classic Illusion Understanding in Vision-Language Models
Junli Zha, Jiahui Wang, Xinkai Lu +1
Vision-Language Models (VLMs) exhibit systematic bias toward visual illusions, recalling memorized facts rather than perceiving actual visual differences. This paper presents a tra…
VAMPO: Policy Optimization for Improving Visual Dynamics in Video Action Models
Zirui Ge, Pengxiang Ding, Baohua Yin +16
Video action models are an appealing foundation for Vision--Language--Action systems because they can learn visual dynamics from large-scale video data and transfer this knowledge…
MAGIA: Sensing Per-Image Signals from Single-Round Averaged Gradients for Label-Inference-Free Gradient Inversion
Zhanting Zhou, Jinbo Wang, Zeqin Wu +1
We study gradient inversion in the challenging single round averaged gradient SAG regime where per sample cues are entangled within a single batch mean gradient. We introduce MAGIA…
How to Defend Against Large-scale Model Poisoning Attacks in Federated Learning: A Vertical Solution
Jinbo Wang, Ruijin Wang, Fengli Zhang
Federated learning (FL) is vulnerable to model poisoning attacks due to its distributed nature. The current defenses start from all user gradients (model updates) in each communica…