UNITES Lab

Publication

Our group actively publishes in the fields of machine learning, computer vision, natural language processing, and AI for science. For an up-to-date full paper list, please visit Dr. Chen's Google Scholar.

Quick Navigation: Efficient Deep Learning | Trustworthy AI | Large Language Models | AI for Science | Computer Vision & Multimodal

Efficient Deep Learning & Sparse Neural Networks

[NeurIPS'25] Mozart: Modularized and Efficient MoE Training on 3.5D Wafer-Scale Chiplet Architectures
S Luo*, Y Han*, P Li*, J Qin*, J Peng, YK Zhao, Y Cao, T Chen
[ICML'25] Occult: Optimizing Collaborative Communication across Experts for Accelerated Parallel MoE Training and Inference
S Luo, P Li, J Peng, H Wang, Y (Katie) Zhao, Y (Kevin) Cao, Y Cheng, T Chen
[ICML'25] I2MoE: Interpretable Multimodal Interaction-aware Mixture-of-Experts
J Xin, S Yun, J Peng, I Choi, JL Ballard, T Chen, Q Long
[NAACL'25 SAC Award] Advancing MoE Efficiency: A Collaboration-Constrained Routing (C2R) Strategy for Better Expert Parallelism Design
M Zhang*, P Li*, J Peng, M Qiu, T Chen
[ICLR'25] PortLLM: Personalizing Evolving Large Language Models with Training-Free and Portable Model Patches
R Shahroz Khan, P Li*, S Yun*, Z Wang, S Nirjon, CW Wong, T Chen
[ICLR'25] Cut the Crap: An Economical Communication Pipeline for LLM-based Multi-Agent Systems
G Zhang, Y Yue, Z Li, S Yun, G Wan, K Wang, D Cheng, JX Yu, T Chen
[EMNLP'25 Findings] Bag of Tricks for Sparse Mixture-of-Experts: A Benchmark Across Reasoning, Efficiency, and Safety
M Qiu, Z Shen, P Li, A Li, T Chen
[EMNLP'25 Findings] ORAL: Prompting Your Large-Scale LoRAs via Conditional Recurrent Diffusion
R Khan, D Tang, P Li, K Wang, T Chen
[NeurIPS'24] Model-GLUE: Democratized LLM Scaling for A Large Model Zoo in the Wild
X Zhao*, G Sun*, R Cai*, Y Zhou*, P Li*, P Wang*, B Tan, Y He, L Chen, Y Liang, B Chen, B Yuan, H Wang, A Li, Z Wang, T Chen
[EMNLP'24] FFN-SkipLLM: A Hidden Gem for Autoregressive Decoding with Adaptive Feed Forward Skipping
A Jaiswal, B Hu, L Yin, Y Ro, S Liu, T Chen, A Akella
[EMNLP'24] Is C4 Dataset Optimal for Pruning? An Investigation of Calibration Data for LLM Pruning
A Bandari, L Yin, CY Hsieh, AK Jaiswal, T Chen, L Shen, R Krishna, S Liu
[ICML'24] Revisiting Zeroth-Order Optimization for Memory-Efficient LLM Fine-Tuning: A Benchmark
Y Zhang*, P Li*, J Hong*, J Li*, Y Zhang, W Zheng, PY Chen, JD Lee, W Yin, M Hong, Z Wang, S Liu, T Chen
[ICML'24] Two Heads Are Better Than One: Boosting Graph Sparse Training via Semantic and Topological Awareness
G Zhang, Y Yue, K Wang, J Fang, Y Sui, K Wang, Y Liang, D Cheng, S Pan, T Chen
[ICML'24] Towards Building Reliable Language Models with Sparse Mixture-of-Experts
G Chen, X Zhao, T Chen, Y Cheng
[ICLR'24] Merge, Then Compress: Demystify Efficient SMoE with Hints from Its Routing Policy
P Li, Z Zhang, P Yadav, YL Sung, Y Cheng, M Bansal, T Chen
[ICLR'24] Sparse MoE with Language Guided Routing for Multilingual Machine Translation
X Zhao, X Chen, Y Cheng, T Chen
[AAAI'25] Sparse Transfer Learning Accelerates and Enhances Certified Robustness: A Comprehensive Study
Z Li, T Chen, L Li, B Li, Z Wang
[AAAI'25] Visual Prompting Upgrades Neural Network Sparsification: A Data-Model Perspective
C Jin*, T Huang*, Y Zhang, M Pechenizkiy, S Liu, S Liu, T Chen

Trustworthy AI & Safety

[ACL'25 Oral] Agents Under Siege: Breaking Pragmatic Multi-Agent LLM Systems with Optimized Prompt Attacks
R Khan, Z Tan, S Yun, C Fleming, T Chen
[COLM'25] More is Less: The Pitfalls of Multi-Model Synthetic Preference Data in DPO Safety Alignment
Y Wang, R Chen, B Li, D Cho, Y Deng, R Zhang, T Chen, Z Wang, A Grama, J Hong
[EMNLP'25] Bit-Flip Error Resilience in LLMs: A Comprehensive Analysis and Defense Framework
Y Chen, Z Tan, AK Jaiswal, H Qu, X Zhao, Q Lin, Y Cheng, A Kwong, Z Cao, T Chen
[KDD'25] A Survey on Trustworthy LLM Agents: Threats and Countermeasures
M Yu*, F Meng*, X Zhou, S Wang, J Mao, L Pang, T Chen, K Wang, X Li*, Y Zhang, B An, Q Wen*
[ACL'25 Findings] Unveiling Privacy Risks in Multi-Modal Large Language Models: Task-Specific Vulnerabilities and Mitigation Challenges
T Chen, P Li, K Zhou, T Chen, H Wei
[ACL'25 Findings] Vision Language Model Helps Private Information De-Identification in Vision Data
T Chen, P Li, K Zhou, T Chen, H Wei
[NAACL'25] BPO: Towards Balanced Preference Optimization between Knowledge Breadth and Depth in Alignment
S Wang, Y Tong, H Zhang, D Li, X Zhang, T Chen
[NAACL'25] Layer-Level Self-Exposure and Patch: Affirmative Token Mitigation for Jailbreak Attack Defense
Y Ouyang, H Gu, S Lin, W Hua, J Peng, B Kailkhura, T Chen, K Zhou
[AAAI'25] Tuning-Free Accountable Intervention for LLM Deployment--A Metacognitive Approach
Z Tan, J Peng, T Chen, H Liu
[AAAI'25] Sparsity-Guided Holistic Explanation for LLMs with Interpretable Inference-Time Intervention
Z Tan, T Chen, Z Zhang, H Liu
[EAAI'25] Word-Sequence Entropy: Towards Uncertainty Estimation in Free-Form Medical Question Answering Applications and Beyond
Z Wang, J Duan, C Yuan, Q Chen, T Chen, Y Zhang, R Wang, X Shi, K Xu
[EMNLP'24] "Glue Pizza and Eat Rocks"--Exploiting Vulnerabilities in Retrieval-Augmented Generative Models
Z Tan, C Zhao, R Moraffah, Y Li, S Wang, J Li, T Chen, H Liu
[ICML'24] TrustLLM: Trustworthiness in Large Language Models
Y Huang, L Sun, H Wang, S Wu, Q Zhang, Y Li, C Gao, Y Huang, W Lyu, Y Zhang, et al.
[SDM'25] Protecting Privacy against Membership Inference Attack with LLM Fine-tuning through Flatness
T Chen, L Da, H Zhou, P Li, K Zhou, T Chen, H Wei

Large Language Models & Agents

[ACL'25 Oral] Agents Under Siege: Breaking Pragmatic Multi-Agent LLM Systems with Optimized Prompt Attacks
R Khan, Z Tan, S Yun, C Fleming, T Chen
[EMNLP'25 Oral] EQA-RM: A Generative Embodied Reward Model with Test-time Scaling
Y Chen, Z Tan, T Chen
[EMNLP'25] Task-Aware Resolution Optimization for Visual Large Language Models
W Luo, Z Tan, Y Li, X Zhao, K Lee, B Dariush, T Chen
[EMNLP'25] MedHallu: A Comprehensive Benchmark for Detecting Medical Hallucinations in Large Language Models
S Pandit, J Xu, J Hong, Z Wang, T Chen, K Xu, Y Ding
[EMNLP'25 Findings] ORAL: Prompting Your Large-Scale LoRAs via Conditional Recurrent Diffusion
R Shahroz, D Tang, P Li, K Wang, T Chen
[EMNLP'25 Findings] FIER: Fine-Grained and Efficient KV Cache Retrieval for Long-context LLM Inference
D Wang, Z Liu, S Wang, Y Ren, J Deng, J Hu, T Chen, H Yang
[ICML'25 Spotlight] G-Designer: Architecting Multi-Agent Communication Topologies via Graph Neural Networks
G Zhang, Y Yue, X Sun, G Wan, M Yu, J Fang, K Wang, T Chen, D Cheng
[ACL'25] In Prospect and Retrospect: Reflective Memory Management for Long-term Personalized Dialogue Agents
Z Tan, J Yan, IH Hsu, R Han, Z Wang, LT Le, Y Song, Y Chen, H Palangi, G Lee, A Iyer, T Chen, H Liu, CY Lee, T Pfister
[ACL'25] The Efficiency vs. Accuracy Trade-off: Optimizing RAG-Enhanced LLM Recommender Systems Using Multi-Head Early Exit
H Zhou, H Gu, X Liu, K Zhou, M Liang, Y Xiao, S Govindan, P Chawla, J Yang, X Meng, H Li, B Zhang, L Luo, WY Chen, Y Han, B Long, R Zhang, T Chen
[ACL'25 Findings] UQ-Merge: Uncertainty Guided Multimodal Large Language Model Merging
H Qu, X Zhao, J Peng, K Lee, B Dariush, T Chen
[NAACL'25] GuideLLM: Exploring LLM-Guided Conversation with Applications in Autobiography Interviewing
J Duan*, X Zhao*, Z Zhang*, E Ko, L Boddy, C Wang, T Li, A Rasgon, J Hong, MK Lee, C Yuan, Q Long, Y Ding, T Chen, K Xu
[NeurIPS'24] GTBench: Uncovering the Strategic Reasoning Capabilities of LLMs via Game-Theoretic Evaluations
J Duan*, R Zhang*, J Diffenderfer, B Kailkhura, L Sun, E Stengel-Eskin, M Bansal, T Chen, K Xu
[NAACL'24] ReTA: Recursively Thinking Ahead to Improve the Strategic Reasoning of Large Language Models
J Duan, S Wang, J Diffenderfer, L Sun, T Chen, B Kailkhura, K Xu
[EACL'24] Contextualization Distillation from Large Language Model for Knowledge Graph Completion
D Li, Z Tan, T Chen, H Liu
[ICLR'25] Adapt-∞: Scalable Lifelong Multimodal Instruction Tuning via Dynamic Data Selection
A Maharana*, J Yoon*, T Chen, M Bansal
[COLING'25] Aurora-M: The First Open Source Multilingual Language Model Red-Teamed According to the US Executive Order
T Nakamura, M Mishra, S Tedeschi, Y Chai, JT Stillerman, F Friedrich, P Yadav, T Laud, VM Chien, TY Zhuo, D Misra, B Bogin, XS Vu, M Karpinska, AV Dantuluri, W Kusa, T Furlanello, R Yokota, N Muennighoff, S Pai, T Adewumi, V Laippala, X Yao, A Junior, A Ariyak, A Drozd, J Clive, K Gupta, L Chen, Q Sun, K Tsui, N Persaud, N Fahmy, T Chen, M Bansal, N Monti, T Dang, Z Luo, TT Bui, R Navigli, V Mehta, M Blumberg, V May, H Nguyen, S Pyysalo

AI for Science & Healthcare

[Nature Plants'25] Using Large Language Models to Address the Bottleneck of Georeferencing Natural History Collections
Y Xie, DS Park, MA Sinnott-Armstrong, J Ho, T Chen, AS Weakley, LJ Aguirre Lopez, J Choi, MM Laitinen, NA Steeves, CH Huang, R Xu, X Feng
[ACL'25] MMOFA: A Multi-Omics Foundation Model with Feature-Alignment for Clinical Phenotype Prediction
D Li, Y Yue, R Zhang, Z Tan, T Chen, J Xie, L Bao, D Cheng, H Liu
[KDD'25] Graph Learning under Distribution Shifts: Graph Feature Imputation and Out-of-Distribution Detection
Y Yue, L Yang, Y Li, K Wang, T Chen, Y Sui, D Cheng
[JBI'25] GatorCLR: Personalized Predictions of Patient Outcomes on Electronic Health Records via Similarity-based Contrastive Learning
D Li, C Yuan, R Zhang, N Shang, Q Wei, T Chen, Z Lu, Y Wang
[AAAI'25] Learning to Model the Drift of Biological Neural Network Connectivity from Calcium Imaging Observations
C Jin, Y Xu, J Xiao, T Chen, H Liu, S Liu
[NeurIPS'24] Rethinking Improved Privacy-Utility Trade-off with Pre-existing Knowledge for DP Training
Y Yu, Y Yue, B Chen, G Zhang, Y Liu, A Wei, T Chen, J Gao
[EMNLP'24 Findings] Synergizing Large Language Models and Knowledge Graphs for Alzheimer's Disease Care Plan Generation
D Li, Y Yue, R Zhang, Z Tan, Y Guo, X Gong, T Chen, J Xie, D Cheng, H Liu
[EMNLP'24 Findings] Cross-Lingual Multi-Hop Knowledge Editing -- Benchmarks, Analysis and a Simple Contrastive Learning Based Approach
A Khandelwal*, H Singh*, H Gu, T Chen, K Zhou
[ECCV'24] Mew: Multiplexed Immunofluorescence Image Analysis Through an Efficient Multiplex Network
S Yun, J Peng, AE Trevino, C Park, T Chen
[ICML'24] Evolution-Inspired Loss Functions for Protein Representation Learning
T Bepler, S Liu, T Chen, V Gligorijevic
[Bioinformatics'24] Single-cell RNA Sequencing Data Imputation Using Bi-level Feature Propagation
Z Tang, S Li, X Jiang, S Peng, T Chen, S Liu, J Xu
[CVPR'24] Molecular Data Programming: Towards Molecule Pseudo-labeling with Systematic Weak Supervision
W Zhang, J Hong, Y Wang, T Chen, L Getoor

Computer Vision & Multimodal Learning

[ACM MM'25 Best Paper] VLM-3R: Vision-Language Models Augmented with Instruction-Aligned 3D Reconstruction
Z Fan*, J Zhang*, R Li, J Zhang, R Chen, H Hu, K Wang, H Qu, D Wang, Z Yan, H Xu, J Theiss, T Chen, J Li, Z Tu, Z Wang, R Ranjan
[ICML'25] Rethinking Multimodal Learning from an Optimization Perspective
S Yun, X Zhao, S Chen, T Chen
[ICML'25] UniTok: A Unified Tokenizer for Visual Generation and Understanding
T Huang, L Zhang, C Jin, S Liu, M Pechenizkiy, S Liu, T Chen
[ICLR'25] What Do Learning Dynamics Reveal About Multimodal Representations?
S Yun, I Choi, J Xin, J Peng, JL Ballard, T Chen, Q Long
[NeurIPS'24 Spotlight] Flex: End-to-End Text-Instructed Visual Navigation with Foundation Models
L Harris, K Ni, A Fishman, T Chen, L Paull, N Roy
[AAAI'25] Visual Prompting Upgrades Neural Network Sparsification: A Data-Model Perspective
C Jin*, T Huang*, Y Zhang, M Pechenizkiy, S Liu, S Liu, T Chen
[ECCV'24] Beyond Accuracy: Tracking More Effective Evolutions in Facial Affective Behavior Analysis
Z Wang, Y Chen, K Zhu, B Jiang, Y Han, T Chen, B Zhang
[CVPR'24] TFMQ-DM: Temporal Feature Maintenance Quantization for Diffusion Models
Y Huang*, R Gong*, J Liu, T Chen, X Liu

Awards & Honors

[ACM MM'25] Best Paper Award at MFMSI Workshop
[AMIA-IS'25] Marco Ramoni Distinguished Paper Award
[NAACL'25] Low-Resource Methods for NLP SAC Award
[AAAI'25] Best Paper Award at GenAI4Health Workshop
[NeurIPS'24] Best Demo Paper Award at GenAI4Health Workshop
1st Place of ACM/IEEE Quantum Computing for Drug Discovery Challenge