Xiangjue Dong

xj.dong [at] tamu.edu

Computer Science & Engineering

Texas A&M University

I am a final-year PhD Student in Computer Science at Texas A&M University, advised by Prof. James Caverlee.

My research focuses on AI Agents, LLM Reasoning and Personalization, Trustworthy AI, and Conversational AI. I aim to build personalized language agents that are trustworthy, collaborative, and aligned with human needs — agents capable of reasoning reliably, working collectively with other agents, and adapting to individual users and contexts.

If you would like to chat about life, career, or research ideas related to Agents/LLM/NLP, feel free to reach out.

🔥 What's New

Professional Service

Awards

Personal

Marathon (Personal best: 3:43:15). Bouldering (previous TAMU climbing team). Piano (Grade 10 Certificate, highest honor for amateur pianists).   Strava

* equal contribution; mentored student.   See also Google Scholar.

Recent Preprints

paper
image

CHOIR: Collaborative Harmonization fOr Inference Robustness.

Xiangjue Dong, Cong (Nicole) Wang, Maria Teleki, Millennium Bismay, Ruihong Huang, James Caverlee.

Under review at ACL 2026; presented at NeurIPS 2025 Workshop.

paper
image

DMRetriever: A Family of Models for Improved Text Retrieval in Disaster Management.

Kai Yin, Xiangjue Dong, Chengkai Liu, Allen Lin, Lingfeng Shi, Ali Mostafavi, James Caverlee.

Under review at ACL 2026.

paper
image

DisastQA: A Comprehensive Benchmark for Question Answering Evaluation in Disaster Management.

Zhitong Chen*, Kai Yin*, Xiangjue Dong, Chengkai Liu, James Caverlee.

Under review at ACL 2026.

paper
image

Probing the Limits of Embodied Spatial Planning in LLMs.

Xiangjue Dong, Manling Li, James Caverlee.

NeurIPS 2025 Workshop on Space in Vision, Language, and Embodied AI.

paper
image

A Survey on LLM Inference-Time Improvement.

Xiangjue Dong*, Maria Teleki*, James Caverlee.

Preprint.

paper
image

Probing, Measuring, and Mitigating Gender Affiliations in Large Language Models.

Xiangjue Dong*, Yibo Wang*, Philip S. Yu, James Caverlee.

Preprint.

Publications  

paper
image

Multi-Scale Model Compression via Nested Matrix Learning.

Xiangjue Dong, Aditya Anantharaman, Hemant Pugaliya, Kai Zhong.

The Fifteenth biennial Language Resources and Evaluation Conference (LREC 2026). Oral.

paper
image

Language Models as Semantic Augmenters for Sequential Recommenders.

Mahsa Valizadeh, Xiangjue Dong, Rui Tuo, James Caverlee.

The Fifteenth biennial Language Resources and Evaluation Conference (LREC 2026).

paper
image

Z-Scores: A Metric for Linguistically Assessing Disfluency Removal.

Maria Teleki, Sai Tejas Janjur, Haoran Liu, Oliver Grabner, Ketan Verma, Thomas Docog, Xiangjue Dong, Lingfeng Shi, Cong Wang, Stephanie Birkelbach, Jason Kim, Yin Zhang, James Caverlee.

2026 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2026).

paper
image

DisastIR: A Comprehensive Information Retrieval Benchmark for Disaster Management.

Kai Yin, Xiangjue Dong, Chengkai Liu, Lipai Huang, Yiming Xiao, Zhewei Liu, Ali Mostafavi, James Caverlee.

Findings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP 2025 Findings).

paper
image

A Survey on LLMs for Story Generation.

Maria Teleki, Vedangi Bengali, Xiangjue Dong, Sai Tejas Janjur, Haoran Liu, Tian Liu, Cong Wang, Ting Liu, Yin Zhang, Frank Shipman, James Caverlee.

Findings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP 2025 Findings).

paper
image

ReasoningRec: Bridging Personalized Recommendations and Human-Interpretable Explanations through LLM Reasoning.

Millennium Bismay, Xiangjue Dong, James Caverlee.

Findings of the 2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics (NAACL 2025 Findings).

paper
image

Masculine Defaults via Gendered Discourse in Podcasts and Large Language Models.

Maria Teleki, Xiangjue Dong, Haoran Liu, James Caverlee.

International AAAI Conference on Web and Social Media (ICWSM 2025).

Also presented at IC2S2 2025 (nominated for a methodology award) and SICon@ACL 2025.

paper
image

DA3: A Distribution-Aware Adversarial Attack against Language Models.

Yibo Wang*, Xiangjue Dong*, James Caverlee, Philip S. Yu.

The 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP 2024). Oral.

paper
image

The Neglected Tails in Vision-Language Models.

Shubham Parashar*, Tian Liu*, Zhiqiu Lin*, Xiangjue Dong, Yanan Li, James Caverlee, Deva Ramanan, Shu Kong.

The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2024).

Also accepted at DMLR@ICML 2024 (Oral).

paper
image

Comparing ASR Models in the Context of Speech Disfluencies.

Maria Teleki, Xiangjue Dong, James Caverlee.

Interspeech 2024.

paper
image

Quantifying the Impact of Disfluency on Spoken Content Summarization.

Maria Teleki, Xiangjue Dong, James Caverlee.

Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024).

paper
image

Disfluency Augmented Curriculum Learning for Fluent Text Generation.

Rohan Chaudhury, Maria Teleki, Xiangjue Dong, James Caverlee.

Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024).

paper
image

Everything Perturbed All at Once: Enabling Differentiable Graph Attacks.

Haoran Liu, Bokun Wang, Jianling Wang, Xiangjue Dong, Tianbao Yang, James Caverlee.

The Web Conference 2024 (WWW 2024). Short.

paper
image

Co2PT: Mitigating Bias in Pre-trained Language Models through Counterfactual Contrastive Prompt Tuning.

Xiangjue Dong, Ziwei Zhu, Zhuoer Wang, Maria Teleki, James Caverlee.

Findings of the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP 2023 Findings).

paper
image

PromptAttack: Probing Dialogue State Trackers with Adversarial Prompts.

Xiangjue Dong, Yun He*, Ziwei Zhu*, James Caverlee.

Findings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL 2023 Findings).

Also accepted at TrustNLP Workshop@ACL 2023 and presented at CRA-W 2023.

paper
image

Closed-book Question Generation via Contrastive Learning.

Xiangjue Dong, Jiaying Lu*, Jianling Wang*, James Caverlee.

The 17th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2023). Oral.

paper
image

Weakly Supervised Concept Map Generation through Task-Guided Graph Translation.

Jiaying Lu, Xiangjue Dong, Carl J. Yang.

IEEE Transactions on Knowledge and Data Engineering (TKDE 2023).

paper
image

Howdy Y'all: An Alexa TaskBot.

Majid Alfifi, Xiangjue Dong, Timo Feldman, Allen Lin, Karthic Madanagopal, Aditya Pethe, Maria Teleki, Zhuoer Wang, Ziwei Zhu, James Caverlee.

1st Proceedings of Alexa Prize TaskBot, 2022.

Quarterfinals.

paper
image

Emora: An Inquisitive Social Chatbot Who Cares For You.

Sarah E. Finch, James D. Finch, Ali Ahmadvand, Ingyu (Jason) Choi, Xiangjue Dong, Ruixiang Qi, Harshita Sahijwani, Sergey Volokhin, Zihan Wang, Zihao Wang, and Jinho D. Choi.

3rd Proceedings of Alexa Prize, 2020.

1st Place Winner.

Selected Workshops

paper
image

Probing Explicit and Implicit Gender Bias through LLM Conditional Text Generation.

Xiangjue Dong*, Yibo Wang*, Philip S. Yu, James Caverlee.

Socially Responsible Language Modelling Research Workshop (SoLaR@NeurIPS 2023).

paper
image

Transformer-based Context-aware Sarcasm Detection in Conversation Threads from Social Media.

Xiangjue Dong, Changmao Li, Jinho D. Choi.

ACL Workshop on Figurative Language Processing (FigLang@ACL 2020).

2nd Place Winner.

Research & Industry Positions

Teaching

Research Mentorship