Hua Xu
Associate Professor
Works on multimodal intelligent information processing, including multimodal intent recognition and dialogue, and intelligent optimization.
Hua Xu's lab at Tsinghua works on multimodal & grounded language, dialogue, generation & summarization and vision-language & multimodal. 25 papers since 2023, 1 at top venues, mostly TMM and ICML. 4 PhD students. 20 alumni. Applications close .
At a glance
- papers, 2023–2026
- 7/ year
- at T1 venues
- 4%
- PhD students
- 4
- MS students
- 0
- alumni listed
- 20
Papers by yearfrom the lab homepage, 2023–2026
- 3 papers2026 · T1 0 · collecting
- 11 papers2025 · T1 1
- 4 papers2024 · T1 0
- 7 papers2023 · T1 0
T1 venuesOtherVenues: TMM, ICML
Research focuspaper titles that use each topic's words
- Multimodal & grounded language11 papers
- Vision-language & multimodal11 papers
- multimodal8 papers
- Dialogue, generation & summarization1 paper
- intelligent information0 papers
Paper list
From the lab homepage.
20263 papers
- Proceedings of the ACM International Conference on Multimedia (MM '26), Portland, OR, USA, .Concept-Guided Reinforcement Learning and the ERAS Benchmark for Explainable Multi-modal Sentiment Analysis
- IEEE Transactions on Cybernetics, .HEQP: A Hypergraph Neural Network-Based Evolutionary Method for Large-Scale QCQPs
- Proceedings of the ACM International Conference on Multimedia (MM '26), Portland, OR, USA, .Unsupervised Multimodal Intent Discovery via MLLM-Guided Concept Generation and Semantic Propagation
202511 papers
- Intelligent Systems with Applications, : 200498.A unified prompt-based framework for few-shot multimodal language analysis
- Advances in Neural Information Processing Systems NeurIPS, .Can Large Language Models Help Multimodal Language Analysis? MMLA: A Comprehensive Benchmark
- Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment. .Ellipsoid-Based Decision Boundaries for Open Intent Classification
- —Intelligent Evolution Optimization: Guided from Deep Learning to Large Language Model//Proceedings of the Genetic and Evolutionary Computation Conference 2025
- —Large Language Model Driven Evolutionary Optimization// IEEE Congress on Evolutionary Computation 2025
- ICML (T1)Large Language Model-driven Large Neighborhood Search for Large-Scale MILP Problems
- Intelligent Systems with Applications, : 200497.Learning how to transfer: A lifelong domain knowledge distillation framework for continual MRC
- IEEE Transactions on Evolutionary Computation, .Light-EvoOPT: A Lightweight Evolutionary Optimization Framework for Ultra-Large-Scale Mixed Integer Linear Programs
- Proceedings of the Conference on Empirical Methods in Natural Language Processing. : 22221-22237.LLM-Guided Semantic Relational Reasoning for Multimodal Intent Recognition
- —MILPBench: A Large-scale Benchmark Test Suite for Mixed Integer Linear Programming Problems.Proceedings of the Genetic and Evolutionary Computation Conference
- //IEEE Transactions on MultimediaMultimodal Classification and Out-of-distribution Detection for Multimodal Intent Understanding
20244 papers
- TMMCrossmodal Translation Based Meta Weight Adaption for Robust Image-Text Sentiment Analysis
- Proceedings of the Genetic and Evolutionary Computation Conference Companion. : 159-162.Learning to generate scalable milp instances
- TMMMeta Noise Adaption Framework for Multimodal Sentiment Analysis With Feature Noise
- Brain Sciences, , 15(1): 30.ViT-Based Face Diagnosis Images Analysis for Schizophrenia Detection
20237 papers
- Health Information Science and Systems, , 11(1): 19.A Two-stage Segmentation of Sublingual Veins Based on Compact Fully Convolutional Networks for Traditional Chinese Medicine Images
- Proceedings of the AAAI Conference on Artificial Intelligence. : 16376-16377.Adaptive Constraint Partition Based Optimization Framework for Large-scale Integer Linear Programming (student abstract)
- The Fortieth International Conference on Machine Learning (ICML ).GNN&GBDT-Guided Fast Optimizing Framework for Large-scale Integer Programming
- IEEE Transactions on Intelligent Transportation Systems, .High-Dimensional Multi-Objective Bayesian Optimization With Block Coordinate Updates: Case Studies in Intelligent Transportation System
- TMMNoise Imitation based Adversarial Training for Robust Multimodal Sentiment Analysis
- Proceedings of the AAAI Conference on Artificial Intelligence. : 16458-16460.Robust-MSA: Understanding the Impact of Modality Noise on Multimodal Sentiment Analysis
- Proceedings of the AAAI Conference on Artificial Intelligence. : 16188-16189.Self-Paced learning based Graph Convolutional Neural Network for Mixed Integer Programming(Student Abstract)
Where alumni went20 listed
Apply to TsinghuaFall 2027
Computer Science PhD (international applicants)
- Round
- Main
- Opens
- GRE
- See program page
- English
- Per school catalog
- Fee
- RMB 800
Not yet confirmed for this intake.
Similar labsnearest by research summary
- USUMD
Douglas Oard
- Knowledge, QA & retrieval
- USNYU
Tal Linzen
- Multilingual, translation & linguistics
- CAToronto
Colin Raffel
- Large language models
- SGNTU
Wei Lu
- Large language models
- CNSJTU
Lu Chen
- Large language models
- USMIT
Yoon Kim
- Efficient training & inference
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
Tsinghuacloses