Wei Xu
Associate Professor
Studies natural language processing, including text generation, stylistics, information extraction and writing assistance.
Wei Xu's lab at Georgia Tech works on knowledge, QA & retrieval and dialogue, generation & summarization. 38 papers since 2023, 30 at top venues, mostly EMNLP and ACL. 6 PhD students and 7 master's students. 14 alumni, most often to Apple and Baidu. Applications close .
At a glance
- papers, 2023–2026
- 10/ year
- at T1 venues
- 79%
- PhD students
- 6
- MS students
- 7
- alumni listed
- 14
Papers by yearfrom the lab homepage, 2023–2026
- 13 papers2026 · T1 9 · collecting
- 12 papers2025 · T1 10
- 13 papers2024 · T1 11
- 0 papers2023 · T1 0
T1 venuesOtherVenues: EMNLP, ACL, Findings of EMNLP, ICLR, NeurIPS
Research focuspaper titles that use each topic's words
- Knowledge, QA & retrieval3 papers
- Dialogue, generation & summarization2 papers
- text generation0 papers
- assistance0 papers
- information extraction0 papers
Paper list
From the lab homepage.
202613 papers
- EMNLP (T1)Camellia: Benchmarking Cultural Biases in LLMs for Asian Languages
- Findings of EMNLPCoCoA: Context-Conditional Cultural Alignment for Large Language Models
- ICLR (T1)Do Vision-Language Models Respect Contextual Integrity in Location Disclosure?
- EMNLP (T1)GAVEL: Agent Meets Checklist for Evaluating LLMs on Long-Context Legal Summarization
- ACL (T1)GeoRC: A Benchmark for Geolocation Reasoning Chains
- —ICLR 2026
- EMNLP (T1)Investigating and Alleviating Harm Amplification in LLM Interactions
- ICML (T1)Learning to Route Languages for Multilingual Policy Optimization
- Findings of EMNLPLocalizing Prompt Ambiguity in Large Language Models with Probe-Targeted Attribution
- NeurIPS (T1)Parallel Universes for Temporal Evaluation and Knowledge Updating in LLMs
- NeurIPS (T1)Soft Token Alignment for Cross-Lingual Reasoning
- CHI (T1)Supporting Informed Self-Disclosure: Design Principles for Presenting AI-Estimates of Privacy Risks to Users
202512 papers
- COLMBeyond the Reported Cutoff: Where Large Language Models Fall Short on Financial Knowledge
- NeurIPS (T1)BRANCH: Probabilistic Reasoning with LLMs for Privacy Risk Estimation
- EMNLP (T1)CARE: Multilingual Human Preference Learning for Cultural Awareness
- AAAI (T1)CROSSNEWS: A Cross-Genre Authorship Verification and Attribution Benchmark
- COLMEvaluating LLMs on Chinese Idiom Translation
- ICLR (T1)Generating CAD Code with Vision-Language Models for 3D Designs
- EMNLP (T1)How to Protect Yourself from 5G Radiation? Investigating LLM Responses to Implicit Misinformation
- CSCW (T1)Measuring, Modeling, and Helping People Account for Privacy Risks in Online Self-Disclosures with AI
- NAACL (T1)On The Origin of Cultural Biases in Language Models: From Pre-training Data to Linguistic Phenomena
- EMNLP (T1)SimulatorArena: Are User Simulators Reliable Proxies for Multi-Turn Evaluation of AI Assistants?
- NAACL (T1)The Impact of Visual Information in Chinese Characters
- EMNLP (T1)What are Foundation Models Cooking in the Post-Soviet World?
202413 papers
- EMNLP DemoChatHF: Collecting Rich Human Feedback from Real-time Conversations
- ICLR (T1)Constrained Decoding for Cross-lingual Label Projection
- ACL (T1)FactPICO: Factuality Evaluation for Plain Language Summarization of Medical Evidence
- EMNLP (T1)GPT-4 Jailbreaks Itself with Near-Perfect Success Using Self-Explanation
- EMNLP (T1)Granular Privacy Control for Geolocation with Vision Language Models
- ACL · 🏆 Best Social Impact AwardHaving Beer after Prayer? Measuring Cultural Bias in Large Language Models
- EMNLP (T1)Improving Minimum Bayes Risk Decoding with Multi-Prompt
- ACL (T1)InfoLossQA: Characterizing and Recovering Information Loss in Text Simplification
- EMNLP (T1)MedReadMe: A Systematic Study for Fine-grained Sentence Readability in Medical Domain
- ACL (T1)Meta-Tuning LLMs to Leverage Lexical Knowledge for Generalizable Language Style Understanding
- ACL (T1)NEO-BENCH: Evaluating Robustness of Large Language Models with Neologisms
- EMNLP (T1)ReadMe++: Benchmarking Multilingual Language Models for Multi-Domain Readability Assessment
Where alumni went14 listed, 60% to industry
Industry 6Academia 4
Most often to
- Apple
- Baidu
- Bloomberg AI
Also to
- Stanford University
- Georgia Tech
- USC
Apply to Georgia TechFall 2027
Georgia Tech Computer Science PhD
- Round
- Single round
- Opens
- —
- GRE
- optional
- English
- TOEFL/IELTS
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- SGNTU
Wei Lu
- Large language models
- USCMU
Chenyan Xiong
- Large language models
- USUIUC
Heng Ji
- Knowledge, QA & retrieval
- USMIT
Yoon Kim
- Efficient training & inference
- USNYU
Tal Linzen
- Multilingual, translation & linguistics
- CNTsinghua
Hua Xu
- Multimodal & grounded language
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.
Georgia Techcloses