Page Not Found
Page not found. Your pixels are in another canvas.
A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
Page not found. Your pixels are in another canvas.
About me
This is a page not in th emain menu
Published:
If you need USTC to recruit you, you ain’t the one.
Published:
Published:
Short description of portfolio item number 1
Published:
Short description of portfolio item number 2 
Hanyu Wang, Yujin Han, Difan Zou
Published in ICML 2024 Workshop on Foundation Models in the Wild, 2024
We address the paradox in diffusion models where the theoretical optimal score suggests memorization while trained models generalize, attributing generalization to the superior geometric smoothness of the trained score function.
Hanyu Wang, Bochuan Cao, Yuanpu Cao, Jinghui Chen
Published in Proceedings of the 42nd International Conference on Machine Learning (ICML 2025), 2025
We propose TruthFlow, a representation intervention framework that mitigates LLM hallucination by generating query-specific correction vectors via flow matching, enhanced by a truth-related subspace projection using SVD.
Hanyu Wang, Yuanpu Cao, Lu Lin, Jinghui Chen
Published in Preprint, 2026
We develop a framework that shifts reflection from reactive, post-hoc correction to proactive, pre-execution foresight, anchoring a planning-phase reflection loop with a distilled taxonomy of historical errors.
Yifan Lan, Yuanpu Cao, Hanyu Wang, Lu Lin, Jinghui Chen
Published in Preprint, 2026
We show that chain-of-thought reasoning can mask memorization, and propose the Zero-CoT Probe (ZCP), a black-box method that truncates the reasoning process and compares performance against an isomorphically perturbed reference set to detect both direct and evasive data contamination.
Hanyu Wang, Yifan Lan, Bochuan Cao, Lu Lin, Jinghui Chen
Published in Preprint, 2026
We optimize agent skills in a gradient-descent-like loop, iteratively refining skill documents from execution feedback.
Yifan Lan, Hanyu Wang, Lu Lin, Jinghui Chen
Published in NeurIPS 2026 Workshop on Towards Test-Time Continual Learning Agents (TTCL), 2026
We adapt agent skills authored for strong models so that weaker agents can execute them well, by abstracting strong and weak trajectories into typed execution structures, localizing their divergence with graph edit distance, and iteratively patching the skill.
Hanyu Wang, Nakul Agarwal, Hossein Nourkhiz Mahjoub, Ehsan Moradi Pari, Makoto Fukushima, Jinghui Chen, Vaishnav Tadiparthi
Published in Preprint, 2026
We study how much reference-prefix guidance to use during rollouts in RLVR, derive theoretical bounds on the induced KL divergence, and propose Adaptive Reference Guidance, which improves mathematical reasoning performance.
Yifan Lan, Zhongye Liu, Yurui Chang, Hanyu Wang, Yongkang Du, Lu Lin, Jinghui Chen
Published in Preprint, 2026
We introduce DecisionRobust, a benchmark for evaluating the adversarial robustness of decision-making models across both textual and visual inputs.
Yuanpu Cao, Saket Sathe, Hanyu Wang, Ziyi Yin, Fenglong Ma, Jinghui Chen
Published in Advances in Neural Information Processing Systems (NeurIPS 2026), 2026
We propose Progress-Aware Distillation (ProD), which detects and penalizes unproductive loops in student agents during distillation from larger models, helping small language model agents keep making progress on complex tasks.
Published:
This is a description of your talk, which is a markdown files that can be all markdown-ified like any other post. Yay markdown!
Published:
This is a description of your conference proceedings talk, note the different field in type. You can put anything in this field.
Workshop, University 1, Department, 2015
This is a description of a teaching experience. You can use markdown like any other post.
Undergraduate course, Penn State, Department, 2025
TA