5 papers
From Single-Actor Environments to Multi-Actor Event Worlds: Can LLMs Be Event World Models?
Juntong Ni, Zewen Liu, Bohan Wang +2
World models help agents assess the consequences of their actions before acting. Many real systems, however, involve multiple actors who make their own decisions, can evolve even w…
Agent Policy-Value Audit: Separating Transition Composition from Event Selection in Financial LLM Agents
Mingyang, Chen, Yida +4
Financial LLM agents are often evaluated by comparing their end-to-end returns with those of a baseline and testing the paired difference against zero. This measures whether deploy…
Adaptive Auto-Harness: Sustained Self-Improvement for Agentic System Deployment on Open-Ended Task Streams
Zewen Liu, Zhan Shi, Yisi Sang +7
Auto-harness systems such as A-Evolve, GEPA, and Meta-Harness improve LLM agents by optimizing prompts, skills, tools, memories, and supporting infrastructure from execution feedba…
Exposing Vulnerabilities in Explanation for Time Series Classifiers via Dual-Target Attacks
Bohan Wang, Zewen Liu, Lu Lin +4
Interpretable time series deep learning systems are often assessed by checking temporal consistency on explanations, implicitly treating this as evidence of robustness. We show tha…
Score-based Conditional Out-of-Distribution Augmentation for Graph Covariate Shift
Bohan Wang, Yurui Chang, Wei Jin +1
Distribution shifts between training and testing datasets significantly impair the model performance on graph learning. A commonly-taken causal view in graph invariant learning sug…