3 papers
stat.ML2026
Should the Boundary Term Be Learned in Reflected Diffusion? Conormal Trace and Reflection Masking
Ziyue Wang, Takafumi Kanamori
We study score learning for reflected diffusion on bounded domains. Reflection keeps trajectories feasible but does not ensure that the learned score satisfies the boundary behavio…
cs.AI2026
When Agents Overtrust Environmental Evidence: An Extensible Agentic Framework for Benchmarking Evidence-Grounding Defects in LLM Agents
Strick Sheng, Ziyue Wang, Liyi Zhou
Large language model agents increasingly operate through environment-facing scaffolds that expose files, web pages, APIs, and logs. These observations influence tool use, state tra…
cs.LG2026
Distributional Spectral Diagnostics for Localizing Grokking Transitions
Ziyue Wang, Yufeng Ying, Takafumi Kanamori
In grokking, a model first fits the training data while test accuracy remains low, and only later begins to generalize. We ask whether this transition can be localized from observe…