I am an AI Research Scientist at Meta. Previously, I was a researcher at NEC Laboratories America. I received the Ph.D. in Computer Science Department at The University of Texas at Dallas, in 2022. My advisor is Prof. Feng Chen . I also got an MS at the University of Science and Technology of China (USTC) and a bachelor at Chongqing University in China..
07/2022: I will serve as the PC/reviewer of ICLR,AAAI,KDD,ICML,NeurIPS 2023
01/2022: one paper got accepted by ICASSP 2022
09/2021: one paper got accepted by NeurIPS 2021
08/2021: one paper got accepted by EMNLP 2021
06/2021: I will serve as the PC/reviewer of ICLR,WSDM,AAAI,KDD,ICML,NeurIPS 2022
01/2021: one paper got accepted by WWW 2021
12/2020: one paper got accepted by AAAI 2021
11/2020: I will serve as the PC Member of KDD, NeurIPS 2021
09/2020: two paper got accepted by NeurIPS 2020, one Spotlight, one Poster
Experience
AI Reserach Scientist at Meta (New York, NY) 2026 summer - Current
Reseracher at NEC Laboratories America (Princeton, NJ) 2022 summer - 2026 summer
Reserach intern at NEC Laboratories America (Princeton, NJ) during 2021 summer
Reserach intern at Alibaba Damo Academy (Seattle, WA) during 2019 summer
Research Interests
My research focuses on bridging the gap between foundation models and real-world autonomous decision-making, with frameworks that enable AI agents to solve complex tasks through self-evolution, strategic reasoning, and rigorous reliability guarantees.
Agentic AI: Autonomous Reasoning & Evolution: build next-generation agent frameworks that go beyond static prompting by incorporating long-term memory, strategic search, and reinforcement learning.
LLM Post-Training: Efficiency & Adaptation: make foundation models more efficient and specialized for domain-specific applications while maintaining structural integrity.
Trustworthy AI: Uncertainty & Reliability: quantify what a model doesn't know to ensure safe deployment in high-stakes environments.
Research Publications
†: Corresponding author; *: Equal contribution.
Conference Proceedings
Yutong Cheng, Haifeng Chen, Wenchao Yu, Xujiang Zhao, Peng Gao, Wei Cheng.
"Escaping Whack-a-Mole: Code Documentation Optimization via Dependency-Guided Bi-level Search."
International Conference on Machine Learning. ICML 2026.[Agentic AI][Code Generation]
Linlin Yu, Xujiang Zhao†, Dong Li, Yanchi Liu, Wei Cheng, Zhengzhang Chen, Chen Zhao, Feng Chen, Haifeng Chen.
"Uncertainty-Aware Test-Time Search for Optimization Problem Solving."
In Proceedings of the Association for Computational Linguistics. ACL 2026.[Agentic AI][Test Time Scaling]
Binchi Zhang, Xujiang Zhao, Jundong Li, Haifeng Chen, Zhengzhang Chen.
"Mind the Gap in Cultural Alignment: Task-Aware Culture Management for Large Language Models."
In Proceedings of the Association for Computational Linguistics. ACL 2026.[Post Training]
Xuyuan Liu, Shengyu Chen, Xinshuai Dong, Yanchi Liu, Xujiang Zhao, Haoyu Wang, Yujun Yan, Haifeng Chen, Zhengzhang Chen.
"Representation Interventions Enable Lifelong Unstructured Knowledge Control."
In Proceedings of the Association for Computational Linguistics. ACL 2026.[Post Training][Memory]
Minghao Guo, Qingcheng Zeng, Xujiang Zhao, Yanchi Liu, Wenchao Yu, Mengnan Du, Haifeng Chen, Wei Cheng.
"DeepSieve: Information Sieving via LLM-as-a-Knowledge-Router."
In Findings of the European Chapter of the Association for Computational Linguistics. EACL 2026.[Agentic AI][Workflow]
Wangyang Ying, Yanchi Liu, Xujiang Zhao, Wei Cheng, Zhengzhang Chen, Wenchao Yu, Yanjie Fu, Haifeng Chen.
"Multi-Agent Procedural Graph Extraction with Structural and Logical Refinement."
In Findings of the European Chapter of the Association for Computational Linguistics. EACL 2026.[Agentic AI][Workflow]
Minhua Lin, Zhengzhang Chen, Yanchi Liu, Xujiang Zhao, Zongyu Wu, Junxiang Wang, Xiang Zhang, Suhang Wang, Haifeng Chen.
"Decoding Time Series with LLMs: A Multi-Agent Framework for Cross-Domain Annotation."
In Findings of the European Chapter of the Association for Computational Linguistics. EACL 2026.[Agentic AI][Time Series]
Dong Li, Zhengzhang Chen, Xujiang Zhao, Linlin Yu, Zhong Chen, Yi He, Haifeng Chen, Chen Zhao.
"MARLIN: Multi-Agent Reinforcement Learning for Incremental DAG Discovery."
In Proceedings of the AAAI Conference on Artificial Intelligence. AAAI 2026.[Reinforcement Learning]
Zhixia He, Chen Zhao, Minglai Shao, Xintao Wu, Xujiang Zhao, Dong Li, Qin Tian, Linlin Yu.
"Out-of-Distribution Detection with Positive and Negative Prompt Supervision Using Large Language Models."
In Proceedings of the AAAI Conference on Artificial Intelligence. AAAI 2026.[Trustworthy AI]
Qin Tian, Chen Zhao, Xintao Wu, Dong Li, Minglai Shao, Xujiang Zhao, Wenjun Wang.
"Class-Domain Incremental Learning on Graphs via Disentangled Knowledge Distillation."
In Proceedings of The Web Conference. WWW 2026.[Trustworthy AI]
Qiwei Zhao*, Dong Li*, Yanchi Liu, Wei Cheng, Yiyou Sun, Mika Oishi, Takao Osaki, Katsushi Matsuda, Huaxiu Yao, Chen Zhao, Haifeng Chen, Xujiang Zhao†.
"Uncertainty Propagation on LLM Agent."
In Proceedings of the Association for Computational Linguistics. ACL 2025.[Agentic AI][Reliable]
Jonathan Light, Yue Wu, Yiyou Sun, Wenchao Yu, Yanchi Liu, Xujiang Zhao, Ziniu Hu, Haifeng Chen, Wei Cheng.
"Scattered Forest Search: Smarter Code Space Exploration with LLMs."
In International Conference on Learning Representations. ICLR 2025.[Agentic AI][Code Generation]
Xinyuan Wang, Yanchi Liu, Wei Cheng, Xujiang Zhao, Zhengzhang Chen, Wenchao Yu, Yanjie Fu, Haifeng Chen.
"MixLLM: Dynamic Routing in Mixed Large Language Models."
In Proceedings of the North American Chapter of the Association for Computational Linguistics. NAACL 2025.[Efficient AI]
Chengyuan Deng, Zhengzhang Chen, Xujiang Zhao, Haoyu Wang, Junxiang Wang, Jie Gao, Haifeng Chen.
"Correlation-aware Online Change Point Detection."
In Proceedings of the ACM International Conference on Information and Knowledge Management. CIKM 2025.[Trustworthy AI]
Chen Ling, Xujiang Zhao†, Xuchao Zhang, Wei Cheng, Yanchi Liu, Yiyou Sun, Mika Oishi, Takao Osaki, Katsushi Matsuda, Jie Ji, Guangji Bai, Liang Zhao, Haifeng Chen.
"Uncertainty Quantification for In-Context Learning of Large Language Models."
In Proceedings of the North American Chapter of the Association for Computational Linguistics. NAACL 2024.[Reliable LLM]
Nan Zhang, Yanchi Liu, Xujiang Zhao, Wei Cheng, Runxue Bao, Rui Zhang, Prasenjit Mitra, Haifeng Chen.
"Pruning as a Domain-Specific LLM Extractor."
In Findings of the Association for Computational Linguistics. NAACL 2024.[Post Training][Model Compression]
Yijia Xiao, Yiqiao Jin, Yushi Bai, Yue Wu, Xianjun Yang, Xiao Luo, Wenchao Yu, Xujiang Zhao, Yanchi Liu, Quanquan Gu, Haifeng Chen, Wei Wang, Wei Cheng.
"Large Language Models Can Be Good Privacy Protection Learners."
In Proceedings of the Conference on Empirical Methods in Natural Language Processing. EMNLP 2024.[Post Training][Privacy]
Yujie Lin, Chen Zhao, Minglai Shao, Xujiang Zhao, Haifeng Chen.
"Adaptation Speed Analysis for Fairness-aware Causal Models."
In Proceedings of the ACM International Conference on Information and Knowledge Management. CIKM 2023.[Trustworthy AI]
Xujiang Zhao, Xuchao Zhang, Wei Cheng, Wenchao Yu, Yuncong Chen, Haifeng Chen, Feng Chen.
"SEED: Sound Event Early Detection via Evidential Uncertainty."
In IEEE International Conference on Acoustics, Speech and Signal Processing. ICASSP 2022.[Application]
Shang Ma, Haifeng Chen, Yuanzhou Chen, Yanchi Liu, Xujiang Zhao, Wenchao Yu, Yanfang Ye, Xusheng Xiao, Wei Cheng.
"The Geometry of Fakeness: OOD Learning for AI-Image Detection with Information Bottleneck."
NeurIPS 2026 Under Review.[Trustworthy AI]
Jiaqi Liu, Shi Qiu, Mairui Li, Bingzhou Li, Peng Xia, Siwei Han, Haonian Ji, Letian Zhang, Hardy Chen, Haoqin Tu, Xinyu Yang, Xujiang Zhao, Haifeng Chen, Jiawei Zhou, Xiao Wang, Hongtu Zhu, Yun Li, Jiaheng Zhang, Yuyin Zhou, Sheng Wang, James Zou, Zeyu Zheng, Cihang Xie, Mingyu Ding, Huaxiu Yao.
"AutoResearchClaw: End-to-end Autonomous Research From Idea to Paper."
NeurIPS 2026 Under Review.[Agentic AI][Self Evolving]
Hongyu Cao, Yanchi Liu, Kunpeng Liu, Xujiang Zhao, Wei Cheng, Yanjie Fu, Haifeng Chen.
"Data-centric Small LLM Learning: A Gradient Admission Perspective."
NeurIPS 2026 Under Review.[Agentic AI][Memory]
Xinshuai Dong, Haifeng Chen, Xuyuan Liu, Shengyu Chen, Haoyu Wang, Yanchi Liu, Xujiang Zhao, Kun Zhang, Zhengzhang Chen.
"TaPE: Certified Robustness Against Table Permutations with Tabular Positional Encoding."
NeurIPS 2026 Under Review.[Post Training]
Bangwei Guo, Xujiang Zhao†, Yanchi Liu, Wei Cheng, Shengyu Chen, Dongyue Li, Morimoto Masaharu, Takayuki Kuroda, Dimitris N. Metaxas, Haifeng Chen.
"TopoAgent: A Structure-Aware Perception-to-Reasoning Framework for Diagram-to-Graph Topology Extraction with Large Vision-Language Models."
EMNLP 2026 Under Review.[Agentic AI][Vision Reasoning]
Xinyu Wu, Yanchi Liu, Dong Li, Xujiang Zhao, Wei Cheng, Zhengzhang Chen, Chen Zhao.
"FlexTag: Instruction Tagging for Multi-Perspective Understanding with Small Language Models."
EMNLP 2026 Under Review.[Post Training]
We theoretically analyze the failure case and reason for treating LLM detection tasks as binary classification tasks and propose to transform the task to an out-of-distribution detection task.
We present a comprehensive survey on domain specification techniques for large language models, an emerging direction critical for large language model applications. This paper was cited by 2024 Economic Report of the President of the United States
In this paper, we have shown that framing code generation as an optimization task over the code space and applying SCATTERED FOREST SEARCH is highly effective..
In this paper, we proposed MixLLM, a dynamic routing system that selects the most suitable LLM for each query by balancing response quality, cost, and latency.
In this paper, we first derive a unified view of prompt tuning and then present a novel dynamic prompting approach that can significantly improve the performance of prompt tuning while adding only a few additional parameters..
In this paper, we introduce a novel evidential method, Multi-Label Evidential Graph Neural Networks, to predict uncertainty for multiple classes on graph data.
In this paper, we present a comprehensive exploration of strategies for fine-tuning Large Language Models (LLMs) to incorporate domain-specific knowledge while upholding data privacy.
We introduce D-PRUNER, an innovative unstructured dual-pruning method for domain-specific compression on LLM. It is able to extract a compressed, domain-specific, and task-agnostic LLM by identifying weights that are pivotal for both generality and specificity.
We present an off-the-shelf framework KEEP to predict answers for open-ended commonsense reasoning without requiring answer candidates and a pre-defined answer scope.
We study the problem of early event detection in multi-lable classification, and propose a novel framework, Multi-Label Temporal Evidential Neural Network (MTENN), for multi-label uncertainty estimation in temporal data.
In this survey paper, we provide a comprehensive technical review of the existing knowledge-enhanced reasoning techniques across the diverse range of application domains.
In this survey paper, we study the mature uncertainty research in belief/evidence theories in machine learning/deep learning to tackle complex problems under different types of uncertainty.
We study the key causes about the negative impact of OODs (boundary OODs and faraway ODDs) on SSL and proposed a simple unified robust SSL approach for many existing SSL algorithms in order to improve their robustness against OODs.
We propose a novel Polyphonic Evidential Neural Network to model the evidential uncertainty of the class probability with Beta distribution to solve the sound event early detection problem.
We propose a subset selection algorithm in semi-supervised learning to speed up the SSL training. In addition, this algorithm achieve a better performance when unlabeled data consists of Out-of-Distribution (OOD) data and imbalance.
In this work, we propose a self-learning framework that further utilizes unlabeled data of target languages, combined with uncertainty estimation in the process to select high-quality silver labels..
By considering the multidimensional uncertainty, we proposed a novel uncertainty-aware evidential NN called WGAN-ENN (WENN) for solving an out-of-distribution (OOD) detection problem.
A multi-source uncertainty framework of GNN that reflecting various types of uncertainties in both deep learning and belief/evidence theory domains for node classification predictions.
we propose a highly scalable opinion inference probabilistic model, namely Adversarial Collective Opinion Inference (Adv-COI), which provides a solution to infer unknown opinions with high scalability and robustness under the presence of uncertain,.
This paper presents a new approach, called a regularized EEvidential Neural Networks (ENN), that learns an ENN based on regularizations related to different characteristics of inherent data uncertainty.
This paper proposed three DRL-based schemes combining Subjective Logic and deep reinforcement learning where a reward is given based on a different type of uncertainty (i.e., vacuity, dissonance, or monosonance).
GCN-GRU:Combine Graph Convolutional Network (GCN) and the Gated Recurrent Units (GRU) to model the topological and temporal heterogeneous dependency information of a given dynamic network and conflicting opinions based on robust statistics (uncertainty)
GCN-VAE-SL: The proposed DL-based opinion inference model handles node-level opinions explicitly in a large-scale network using graph convoluational network (GCN) and variational autoencoder (VAE) techniques.
We propose a general framework to model and infer heterogeneous uncertainty information in network data based on GCN and node-level opinions via knowledge distillation.
Service
Program Committee Member
2025: ICLR, NeurIPS, ICML, AAAI, ARR, IEEE Transactions on Medical Imaging, ACM Computing Surveys
