Publications

Conference Papers

Counterfactual Self-Evolving Agents for Evidence-Grounded Reasoning

Published in arXiv preprint, 2026

We introduce counterfactual self-evolution, in which a trainable proposer constructs targeted evidence edits and causal explanations, while accepted counterfactuals accumulate in memory to improve a frozen solver across clinical, fact-verification, and business reasoning tasks.

Recommended citation: Xing Han, Yuxin Wang, Chen Chen, Wei Dai, Gautham Krishna Gudur, Shijun Li, Hsing-Huan Chung, Gregory D. Hager, Joydeep Ghosh, Paul Pu Liang, Suchi Saria. (2026). "Counterfactual Self-Evolving Agents for Evidence-Grounded Reasoning." arXiv preprint arXiv:2609.32870.
Download Paper

Unveiling Multi-regime Patterns in SciML: Distinct Failure Modes and Regime-specific Optimization

Published in International Conference on Machine Learning (ICML), 2026

We study multi-regime behavior in scientific machine learning (SciML) models through a regime-aware diagnostic framework that jointly analyzes performance, training dynamics, and loss-landscape geometry. We identify a consistent three-regime structure across PINNs, Neural Operators, and Neural ODEs, show that optimization effectiveness is regime-specific, and uncover fine-grained failure modes (e.g., deceptive sharpness and flatness) that challenge conventional loss-landscape interpretations.

Recommended citation: Yuxin Wang, Yuanzhe Hu, Xiaokun Zhong, Xiaopeng Wang, Haiquan Lu, Tianyu Pang, Michael W. Mahoney, Yujun Yan, Pu Ren, Yaoqing Yang. (2026). "Unveiling Multi-regime Patterns in SciML: Distinct Failure Modes and Regime-specific Optimization." ICML 2026.
Download Paper