Efficient Reasoning

CyclicReflex: Improving Large Reasoning Models via Cyclical Reflection Token Scheduling

Chongyu Fan, Yihua Zhang, Jinghan Jia, Alfred Hero, Sijia Liu

[Paper] [Code]

EPiC: Towards Lossless Speedup for Reasoning Training through Edge-Preserving CoT Condensation

Jinghan Jia, Hadi Reisizadeh, Chongyu Fan, Nathalie Baracaldo, Mingyi Hong, Sijia Liu

[Paper] [Code]

Machine Unlearning

Reasoning Model Unlearning: Forgetting Traces, Not Just Answers, While Preserving Reasoning Skills

Changsheng Wang*, Chongyu Fan*, Yihua Zhang, Jinghan Jia, Dennis Wei, Parikshit Ram, Nathalie Baracaldo, Sijia Liu

[Paper] [Code]

Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond

Chongyu Fan*, Jinghan Jia*, Yihua Zhang, Anil Ramakrishna, Mingyi Hong, Sijia Liu

ICML 2025 [Paper] [Code]

Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearning

Chongyu Fan*, Jiancheng Liu*, Licong Lin*, Jinghan Jia, Ruiqi Zhang, Song Mei, Sijia Liu

[Paper] [Code]

UnlearnCanvas: A Stylized Image Dataset to Benchmark Machine Unlearning for Diffusion Models

Yihua Zhang, Chongyu Fan, Yimeng Zhang, Yuguang Yao, Jinghan Jia, Jiancheng Liu, Gaoyuan Zhang, Gaowen Liu, Ramana Rao Kompella, Xiaoming Liu, Sijia Liu

NeurIPS 2024 [Paper] [Code] [Website]

Defensive Unlearning with Adversarial Training for Robust Concept Erasure in Diffusion Models

Yimeng Zhang, Xin Chen, Jinghan Jia, Yihua Zhang, Chongyu Fan, Jiancheng Liu, Mingyi Hong, Ke Ding, Sijia Liu

NeurIPS 2024 [Paper] [Code]

Challenging Forgets: Unveiling the Worst-case Forget Sets in Machine Unlearning

Chongyu Fan*, Jiancheng Liu*, Alfred Hero, Sijia Liu

ECCV 2024 [Paper] [Code]

SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation

Chongyu Fan*, Jiancheng Liu*, Yihua Zhang, Dennis Wei, Eric Wong, Sijia Liu

ICLR 2024 Spotlight [Paper] [Code] [Website]