news

group news, paper acceptances, releases and service.

Jul 28, 2026 4 papers have been accepted by COLM 2026, 2 papers have been accepted by ECCV 2026.
Jul 27, 2026 Schoenfeld’s Anatomy of Mathematical Reasoning by Language Models has been selected as “SAC Highlight” on ACL 2026.
Jun 28, 2026 Item Difficulty Modeling Using Fine-tuned Small and Large Language Models won 2026 Best Research Award from the International EAssessment Association (eAA).
May 28, 2026 6 ICML (1 Oral + 1 Spotlight) + 5 ACL (1 Oral) + 5 ICLR + 1 ACM CAIS + 1 EACL + 1 AAAI have been accepted in 2026.
Mar 28, 2026 I will serve as the Area Chair for NeurIPS 2026 and most ARR circles (ACL, EMNLP, etc.).
Jan 28, 2026 I am visiting MBZUAI, UAE as a Visiting Assistant Professor for one year and have experienced the 2026 Middle East Conflict.
Jan 27, 2026 Congratulations to Ming Li on being awarded 2026 Apple Scholar in AI/ML!
Nov 28, 2025 I am going to serve as an Area Chair of ICML 2026.
Sep 28, 2025 4 works (ColorBench: VLMs struggle on colors, WALL-E: neurosymbolic world model, VideoHallu: VLMs hallucinate on synthetic videos, DTO: dynamic thinking pattern optimization) have been accepted by NeurIPS 2025.
Sep 27, 2025 Congratulations to Dongping Chen on being awarded UMD Flagship Fellowship (the only CS recipient this year)!
Aug 28, 2025 3 works (Demos’ Position in Prompt bias, Schoenfeld’s Episode Theory for LLM reasoning, NoWait reasoning on LLMs/VLMs) have been accepted by EMNLP 2025.
Aug 27, 2025 I am going to serve as an Area Chair of ICLR 2026 and AAAI 2026.
Jul 28, 2025 3 works (C3PO, MiP-Overthinking, DynaSaur) have been accepted by COLM 2025.
May 28, 2025 We release BLIP3-o, a fully-open sourced framework to unify image reasoning + generation in one model. It aligns the two fundamental tasks via CLIP-level features. You can find report and demo.
May 27, 2025 6 works (3 main with 2 Oral + 3 findings) have been accepted by ACL 2025.
May 26, 2025 3 works have been accepted by ICML: R2-T2 (Test-Time Multimodal MoE), Preference Controllable Multi-Objective RL, and Tilted Sharpness-Aware Minimization.
Apr 28, 2025 We release WALL-E 2.0 (code) to improve the world model for LLM-based embodied agents. TL;DR: WALL-E 2.0 = Neuro-Symbolic World Model + MPC-based LLM Agent, where Neuro-Symbolic World Model = LLM + Complementary Symbolic knowledge (action rules, scene graph, knowledge graph) extracted from LLM errors in experiences.
Apr 27, 2025 We release ColorBench, the first comprehensive benchmark exposing the weaknesses of existing VLMs on color perception, reasoning, and robustness, across >5,800 images from 11 tasks based on real applications including shopping, medical test-kit reading, map-reading, agriculture, art, wildlife research, etc.
Apr 26, 2025 We release C3PO and R2-T2, which develop a novel class of test-time optimization approaches that can significantly boost the performance of existing MoE VLMs and MoE LLMs by 7-15% on challenging tasks.
Apr 25, 2025 We release two works investigating how training data for LLMs (slow vs. fast thinking, instruction vs. reasoning) lead to different layer-wise gradient patterns and training dynamics, from which we unified various existing data quality metrics into one: small but diverse gradient directions indicate better data.
Apr 24, 2025 We release ATLaS and EEF that study better usage of expert trajectories in supervised/reinforcement finetuning of LLM agents. ATLaS selects only 30% critical steps from expert trajectories to achieve better generalization across tasks/envs, while EEF show that expert failed trajectories can considerably improve the agent exploration in RL.
Apr 23, 2025 I am going to serve as an Area Chair of NeurIPS 2025.
Mar 28, 2025 Florence-VL (3B/8B pretrained/SFT VLMs) trained with depth-breadth fusion has been accepted by CVPR 2025 and the complete training recipe has been open-sourced.
Feb 28, 2025 Check the very first “Aha moment” of multimodal reasoning by RL achieved by VisualThinker-R1-Zero from our TurningPoint-AI team!
Jan 28, 2025 7 ICLR + 3 NAACL accepted, featuring our latest works on synthetic data for post-training, MoE (ICLR Oral), many-objective optimization, in-context transferality, multi-modality imbalance & alignment, oversensitiveness & controllability of GenAI.
Dec 28, 2024 I am going to serve as an Area Chair for ARR Dec 2024 & Feb 2025 (ACL 2025).
Nov 28, 2024 I am going to serve as an Area Chair (SPC) for IJCAI 2025.
Sep 28, 2024 Five papers (3 main + 2 findings) have been accepted by EMNLP 2024.
Sep 27, 2024 I am going to serve as an Area Chair of ICLR 2025.
Jul 28, 2024 We launched TurningPoint AI, a research team from multiple universities and industry (UMD+UCLA+PSU+Google) investigating Multimodal Agents, with the goals of building Trustworthy Embodied AI, Self-Evolving Machines, Compositional Agents, and Controllable AIGC. We already released 8 projects with several ICML and ECCV publications and new datasets.
Jul 27, 2024 2 papers of diffusion models (analysis of negative prompts, extracting discriminative features from generative models) have been accepted by ECCV 2024.
May 28, 2024 4 ICLR + 4 ICML + 6 ACL + 2 NAACL + 2 CVPR have been accepted, featuring our works on controllable AIGC, personalized AI, data-efficient training of LLMs, RLHF, prompt optimization, multi-modal hallucinations, multi-modal and embodied agent, and curriculum reinforcement learning.
Feb 28, 2024 We release a survey on knowledge distillation of LLMs with GitHub repo.
Nov 28, 2023 I will give a talk “Towards Controllable and Personalized AI Models” at UMD CS department seminar on 11/03.
Oct 28, 2023 We release HallusionBench focusing on the Language Hallucination and Visual Illusion of GPT-4V(ision), Llava-1.5, and other multi-modality models. Analyses and insights can be found in the preprint.
Oct 27, 2023 Data recycling and filtering improves instruction-tuning of LLMs, leading to recycled LLMs outperforming other larger LLMs trained on new data and RLHF. We release Reflection-Tuning preprint, codebase, and the model.
Oct 26, 2023 Two papers (How Many Demonstrations Do You Need for In-context Learning, Merging Mixture-of-Experts into One) have been accepted by EMNLP 2023.
Sep 28, 2023 Two papers (multi-modality model distillation for task adaptation, clustered additive modeling for structured federated learning) have been accepted by NeurIPS 2023.
Jul 28, 2023 Two papers (model-adaptive data augmentation curriculum, subclass balancing for long-tail recognition) have been accepted by ICCV 2023.
Jun 28, 2023 How to efficiently optimize the textual instructions applied to API black-box LLMs (e.g., ChatGPT) for solving downstream tasks? Please check our recent work InstructZero, paper and code have been released.
Jun 27, 2023 Invited talk at Purdue Statistics on “Structured Decentralized Learning”.
Jun 26, 2023 Two papers (Meta-Vote Pruning and Eigensubspace of Temporal-Difference Dynamics) have been accepted by ECML/PKDD 2023.
May 28, 2023 I will teach CMSC-421 on “Introduction to Artificial Intelligence” in Fall 2023.
Apr 28, 2023 Three papers (training dynamics of continual learning, continual RL via sparse prompting, structured cooperative learning) have been accepted by ICML 2023. See you at Hawaii in July!
Apr 27, 2023 One paper about personalization in federated recommendation system has been accepted by IJCAI 2023.
Dec 28, 2022 I will teach CMSC-828A on “Fantastic Machine Learning Paradigms and Where to use Them” in Spring 2023.
Dec 27, 2022 I will serve as an SPC (meta-reviewer) for IJCAI 2023.
Nov 28, 2022 One XAI paper on extracting local reasoning chains for subtasks from neural networks such as ResNet and ViT has been accepted by TMLR.
Oct 28, 2022 One paper (adversarial attacks to Question-Answer models) has been accepted by EMNLP 2022.
Sep 28, 2022 Three papers (adversarial augmentation for continual learning, adversarial augmentation for representation learning, federated learning from pre-trained models) have been accepted by NeurIPS 2022.
Jun 28, 2022 I will serve as an Area Chair for Winter Conference on Applications of Computer Vision (WACV) 2023.
May 28, 2022 Two papers about environment-and-task-curriculum for RL and adversarial augmentation for self-supervised learning have been accpeted by ICML 2022.
Apr 28, 2022 One paper about personalized federated learning has been accpeted by IJCAI 2022 as a long presentation.
Apr 27, 2022 One paper of phrase-level textual adversarial attack with label preservation has been accpeted by NAACL 2022 Findings.
Mar 28, 2022 One paper (Learning to Collaborate in Decentralized Learning of Personalized Models) has been accepted by CVPR 2022.
Feb 28, 2022 One paper (Token Dropping for Efficient BERT Pretraining) has been accepted by ACL 2022.
Jan 28, 2022 Three papers (Pareto Policy Pool for Model-based Offline RL, Diverse Client Selection for Federated Learning, Omni-scale CNNs for Time Series) have been accepted by ICLR 2022.–>.
Dec 28, 2021 One paper of Federated Prototype Learning has been accepted to AAAI 2022.
Nov 28, 2021 I will serve as an SPC for SIGKDD 2022.
Sep 28, 2021 Three papers (1 spotlight for Submodular Partitioning, Curriculum RL and Planning, Class-Disentanglement) have been accepted to NeurIPS 2021. Congratulations to Shuang Ao and Kaiwen Yang for their first paper!
Sep 27, 2021 One paper of sentiment bias has been accepted to EMNLP 2021 (findings).
Aug 28, 2021 I will serve as an SPC for AAAI 2022.
Feb 28, 2021 I am selected as an expert reviewer for ICML 2021.
Jan 28, 2021 One paper of curriculum learning and training dynamics has been accepted to AISTATS 2021.
Jan 27, 2021 Three papers (RoCL for curriculum noisy-label learning, AutoLRS for auto-learning rate schedule, IPN for prototype zero-shot learning) have been accepted to ICLR 2021.
Jan 26, 2021 One paper of knowledge graph completion has been accepted to WWW 2021.
Oct 28, 2020 Selected among the top 10% of high-scoring reviewers for NeurIPS 2020.
Sep 28, 2020 One paper of curriculum learning and training dynamics has been accepted to NeurIPS 2020.
Jun 28, 2020 One paper of curriculum semi/self-supervised learning has been accepted to ICML 2020.