Publications

2025

1. ICLR

   

   Integrating Protein Dynamics into Structure-Based Drug Design via Full-Atom Stochastic Flows

   Xiangxin Zhou, Yi Xiao, Haowei Lin, and 7 more authors

   In The Thirteenth International Conference on Learning Representations (ICLR 2025),

   PDF
2. arXiv

   

   Uni-3DAR: Unified 3D Generation and Understanding via Autoregression on Compressed Spatial Tokens

   Shuqi Lu^*, Haowei Lin^*, Lin Yao^*, and 4 more authors

   arXiv preprint arXiv:2503.16278,

   PDF Code Website
3. arXiv

   

   A Neural Symbolic Model for Space Physics

   Jie Ying^*, Haowei Lin^*, Chao Yue^*, and 7 more authors

   arXiv preprint arXiv:2503.07994,

   PDF Code
4. ICML Spotlight

   

   MCU: An Evaluation Framework for Open-Ended Game Agents

   Xinyue Zheng^*, Haowei Lin^*, Kaichen He, and 5 more authors

   In The Forty-second International Conference on Machine Learning (ICML 2025),

   Abs PDF Code

   The first evaluation framework for open-ended game agents, with a task generator to synthesize infinite open-ended tasks and VLM-based evaluator.

5. ICLR

   

   TFG-Flow: Training-free Guidance in Multimodal Generative Flow

   Haowei Lin^*, Shanda Li^*, Haotian Ye, and 4 more authors

   In The Thirteenth International Conference on Learning Representations (ICLR 2025),

   Abs PDF Code

   Adapting training-free guidance method from continuous diffusion to multimodal flow matching (discrete + continuous).

6. Nat. Commun.

   

   A Data-Driven Group Retrosynthesis Planning Model Inspired by Neurosymbolic Programming

   Xuefeng Zhang, Haowei Lin, Muhan Zhang, and 2 more authors

   Nature Communications,

   PDF Code
7. AISTATS

   

   Efficient and Asymptotically Unbiased Constrained Decoding for Large Language Models

   Haotian Ye, Himanshu Jain, Chong You, and 4 more authors

   In The Twenty-Eighth International Conference on Artificial Intelligence and Statistics (AISTATS 2025),

   PDF
8. arXiv

   

   Generative Evaluation of Complex Reasoning in Large Language Models

   Haowei Lin^*, Xiangyu Wang^*, Ruilin Yan^*, and 7 more authors

   arXiv preprint arXiv:2504.02810,

   PDF Code Website
9. Nat. Biomed. Eng.

   

   Peptide Design through Binding Interface Mimicry

   Xiangzhe Kong, Rui Jiao, Haowei Lin, and 6 more authors

   Nature Biomedical Engineering,

   PDF Code

2024

1. TPAMI

   

   Jarvis-1: Open-World Multi-Task Agents with Memory-Augmented Multimodal Language Models

   Zihao Wang, Shaofei Cai, Anji Liu, and 9 more authors

   IEEE Transactions on Pattern Analysis and Machine Intelligence,

   DOI PDF Code
2. arXiv

   

   RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Horizon Generation

   Zihao Wang, Anji Liu, Haowei Lin, and 3 more authors

   arXiv preprint arXiv:2403.05313,

   PDF Code
3. NeurIPS Spotlight

   

   TFG: Unified Training-Free Guidance for Diffusion Models

   Haotian Ye^*, Haowei Lin^*, Jiaqi Han^*, and 6 more authors

   In Advances in Neural Information Processing Systems 37 (NeurIPS 2024),

   Abs PDF Code

   The first unified framework for training-free guidance in diffusion models and comprehensive benchmarks on image, audio, molecules.

4. NeurIPS

   

   OmniJarvis: Unified Vision-Language-Action Tokenization Enables Open-World Instruction Following Agents

   Zihao Wang, Shaofei Cai, Zhancun Mu, and 7 more authors

   In Advances in Neural Information Processing Systems 37,

   PDF
5. ICML

   

   Selecting Large Language Model to Fine-tune via Rectified Scaling Law

   Haowei Lin^*, Baizhou Huang^*, Haotian Ye^*, and 7 more authors

   In The Forty-first International Conference on Machine Learning (ICML 2024),

   Abs PDF Website

   The first paper to introduce the problem of LLM selection. Proposed to use scaling law for LLM selection. Rectified vanilla power law in LLM fine-tuning scaling.

6. ICLR

   

   GROOT-2: Weakly Supervised Multi-Modal Instruction Following Agents

   Shaofei Cai, Bowei Zhang, Zihao Wang, and 4 more authors

   In The Twelfth International Conference on Learning Representations (ICLR 2024),

   PDF
7. ICLR

   

   Class Incremental Learning via Likelihood Ratio-Based Task Prediction

   Haowei Lin, Yijia Shao, Weinan Qian, and 3 more authors

   In The Twelfth International Conference on Learning Representations (ICLR 2024),

   Abs PDF Code

   A theory-guided class incremental learning method, being state-of-the-art for more than 3 years (almost reaching performance upper bound).

2023

1. EMNLP

   

   FLatS: Principled Out-of-Distribution Detection with Feature-Based Likelihood Ratio Score

   Haowei Lin and Yuntian Gu

   In The 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP 2023),

   PDF Code
2. ICLR

   

   Continual Pre-Training of Language Models

   Zixuan Ke^*, Yijia Shao^*, Haowei Lin^*, and 3 more authors

   In The Eleventh International Conference on Learning Representations (ICLR 2023),

   Abs PDF Code Poster

   The first paper to introduce the concept of continual pre-training (CPT), with comprehensive benchmarks and a new method motivated by model sparsity.

2022

1. EMNLP

   

   Continual Training of Language Models for Few-Shot Learning

   Zixuan Ke, Haowei Lin, Yijia Shao, and 3 more authors

   In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP 2022),

   PDF Code Poster
2. EMNLP

   

   Adapting a Language Model While Preserving Its General Knowledge

   Zixuan Ke, Yijia Shao, Haowei Lin, and 3 more authors

   In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP 2022),

   PDF
3. ECML

   CMG: A Class-Mixed Generation Approach to Out-of-Distribution Detection

   Mengyu Wang, Yijia Shao, Haowei Lin, and 2 more authors

   In Joint European Conference on Machine Learning and Knowledge Discovery in Databases,

   PDF