news
| Sep 30, 2026 | ๐ Check out our new papers accepted at the NeurIPS 2026, Sydney, Australia ๐ฆ๐บ. The papers cover: ย ย (i) spatial and multi-view geometric reasoning in VLMs, ย ย (ii) safety alignments in LLMs, ย ย (iii) how reinforcement learning shapes the behavior and reasoning capabilities of LLMs during post-training. These works support our broader goal of data- and model-efficient multimodal learning, through better structure and adaptation. All codes will be released soon ๐ โ stay tuned! |
|---|---|
| Sep 05, 2026 | ๐ Check out our two papers accepted at the Conference on Robot Learning (CoRL 2026), Austin, Texas US ๐บ๐ธ:
|
| Jun 15, 2026 | ๐ Happy to receive an ๐ข๐๐๐๐๐ฎ๐ป๐ฑ๐ถ๐ป๐ด ๐ฅ๐ฒ๐๐ถ๐ฒ๐๐ฒ๐ฟ Award from the CVPR 2026 Organizing Committee! |
| Jun 01, 2026 | ๐ Short versions of (i) Struct-SAM and (ii) SparseSAM have been accepted at AdaptFM@ICML 2026. Another work on (iii) Structural Bottleneck Reasoning of Medical VQA reasoning (Link) has also accepted at FM4LS@ICML 2026. |
| May 14, 2026 | ๐ Honored to receive a Gold Reviewer Award from the ICML 2026 Organizing Committee! |
| May 12, 2026 | ๐ Check out our latest works accepted at (ICML 2026), South Korea ๐ฐ๐ท. They include (i) FOCA, a data-efficient vision-language-action model by learning future predicted world in latent space and naturally supports action-free co-training withy synthetic rollouts from video world model; (ii) Token-level Bregman Preference Optimization, a novel Direct Preference Optimization (DPO) method that enables modelling preferences over next-token actions conditioned on state rather than complete output sequences. Codes will be released soon! |
| Feb 01, 2026 | ๐ Happy to share our latest work Slot-VLA, accepted at ICRA 2026 in Vienna, Austria ๐ฆ๐น ๐. We show that objectโrelationโcentric slot representations enable compact, interpretable, and efficient multi-task robotic manipulation, drastically reducing token complexity while maintaining strong performance and generalization. |
| Jan 26, 2026 | ๐ Our work, namely FACET, which introduces scalable structure-aware and fragment-level modeling via Graph Transformers for molecular learning, has been accepted to ICLR 2026 ๐ง๐ท! |
| Jan 06, 2026 | ๐ The DuFal paper has been accepted at Transactions on Machine Learning Research (TMLR) 2026 and be awarded with a J2C Certification. We will present the paper at the International Conference on Machine Learning (ICML) in July 2026, South Korea ๐ฐ๐ท. |
| Dec 04, 2025 | ๐ Our new work, Dual-Frequency-Aware Learning for High-Fidelity Extremely Sparse-View CBCT Reconstruction, has been accepted (with minor revision) to Transactions on Machine Learning Research (TMLR) 2025. Check it out (here)! |
| Nov 08, 2025 | ๐ Exciting News! Weโre thrilled to share that our two recent works have been accepted to AAAI 2026 in Singapore ๐ธ๐ฌ โ one as an oral and the other as a poster presentation! ๐ i. Multi-Mood โ a multi-modal large language model that integrates video, audio, and text with psychological criteria through reinforcement learning to enable trustworthy and emotionally aligned responses. ii. LIBERO-Mem โ a non-Markovian task suite for short- and long-horizon object tracking and manipulation, featuring temporally sequenced subgoals that challenge models to reason beyond the current observation. ๐ Codes will be released soon ๐ โ stay tuned! |
| Sep 26, 2025 | ๐ Excited to share that our works on (i) ExGra-Med โ a data-efficient multimodal large language model (LLM) for healthcare; (ii) Token Redundancy in 3D Point Cloud Transformers โ uncovering how existing 3D transformers (e.g., Ptv-3, Sonata) are over-tokenized, and proposing an efficient token merging strategy that reduces computation by up to 90-95% while preserving accuracy; and (iii) Over-Optimization in RLHF for LLM Post-Training โ exploring how reinforcement learning from human feedback can lead to alignment instability and proposing new insights into optimization LLM post-training have been accepted to NeurIPS 2025 ๐. Excited to present and discuss them at San Diego ๐บ๐ธ ๐ |
| Sep 09, 2025 | ๐ Excited to give a talk about my current research on Scaling Multi-Modal Learning: Hybrid Representations and Efficient Adaptation at Machine Learning Lab, School of Information and Communications Technology (SOICT), Hanoi University of Science and Technology, Vietnam and (ii) School of Computing, National University of Singapore (NUS). |
| Sep 02, 2025 |  The MGPath has been accepted to the Transactions on Machine Learning Research. Congratulations to all co-authors on this milestone! |
| May 01, 2025 | ๐ Our first (i) preliminary version, MGPath has been accepted to the Workshop on Foundation Models in the Wild, ICLR 2025 and (ii) another one about LLaMA-Adapterโs prompt learning is accepted at ICML 2025. |
| Apr 20, 2025 | ๐ Our work in building a new Inductive Message Passing Network for Efficient Human-in-the-Loop Annotation of Mobile Eye Tracking Data has been accepted at Scientific Report, Nature Portfolio. |
| Feb 20, 2025 | Excited to share our latest work! ๐: (i) On Zero-Initialized Attention: Optimal Prompt and Gating Factor Estimation โ We introduce a Mixture of Experts (MoE) perspective to explain the mechanism behind LLaMA-Adapterโs prompt learning. (ii) MGPath โ A novel multi-granular prompt learning method for few-shot WSI pathology prediction, leveraging the power of foundation vision-language models. |
| Oct 08, 2024 | ๐จ๐ญ Start my visiting research at ETH AI Center, ETH Zurich. The topics are about Multi-Modal LLMs for Healthcare empowered by Retrieval-Augmented Generation. |
| Oct 07, 2024 | Excited to introduce our latest work on medical multi-modal LLMs: LoGra-Med, a novel pre-training algorithm that incorporates multi-graph alignment to effectively address the data-hungry nature of autoregressive learning. |
| Oct 06, 2024 |  The paper PiToMe has been accepted at NeurIPS 2024. Our code will be available soon! |
| Jun 10, 2024 | Our new preprint PiToMe is online. We propose a new method to do token merging in the Transformer with spectrum-preserving. |
| May 01, 2024 | A paper submitted to ICML 2024 on the molecular conformer aggregation network topic is accepted. |
| Jan 15, 2024 | A paper submitted to ICLR 2024 on the topic of accelerating transformers is accepted as an oral talk. |
| Sep 22, 2023 | A paper submitted to NeurIPS 2023 on a large-scale medical image pre-trained models using second-order graph matching is accepted. |