Jieyu Zhao
WiSE Gabilan Assistant Professor of Computer Science
Studies fairness, accountability and trustworthiness of NLP models and LLMs.
Jieyu Zhao's lab at USC works on interpretability, evaluation & safety of LMs, large language models and trustworthy & safe ML. 37 papers since 2023, 17 at top venues, mostly EMNLP and Findings of EMNLP. 4 PhD students and 3 master's students. 19 alumni, most often to Amazon and ByteDance. Applications close .
At a glance
- papers, 2023–2026
- 10/ year
- at T1 venues
- 46%
- PhD students
- 4
- MS students
- 3
- alumni listed
- 19
Papers by yearfrom the lab homepage, 2023–2026
- 10 papers2026 · T1 4 · collecting
- 10 papers2025 · T1 5
- 10 papers2024 · T1 6
- 7 papers2023 · T1 2
T1 venuesOtherVenues: EMNLP, Findings of EMNLP, EACL, ICLR, ICML
Research focuspaper titles that use each topic's words
- Large language models19 papers
- Interpretability, evaluation & safety of LMs2 papers
- Trustworthy & safe ML2 papers
- accountability0 papers
- studies fairness0 papers
Paper list
From the lab homepage.
202610 papers
- ICLR (T1)CoAct-1: Computer-using Agents with Coding as Actions (opens in a new tab)
- ICML (T1)Detecting and Filtering Unsafe Training Data via Data Attribution with Denoised Representation (opens in a new tab)
- TMLR RL finetuning wastes compute on prompts the model already aces We schedule training data adaptively - pushing the model toward problems just past its current capability 📄Paper💻CodealignmentEfficient Reinforcement Finetuning via Adaptive Curriculum Learning (opens in a new tab)
- TMLR Text-to-image models trade diversity for fidelity by default We show you can recover demographic and conceptual diversity without sacrificing how good the images look trustworthinessEnhancing Diversity in Text-to-Image Generation Without Compromising Fidelity (opens in a new tab)
- Lifelong Agent Workshop @ ICLRExperiential Reinforcement Learning (opens in a new tab)
- EACLGRAVITY: A Framework for Personalized Text Generation via Profile-Grounded Synthetic Preferences (opens in a new tab)
- ICLR (T1)On the Trustworthiness of Generative Foundation Models: Guideline, Assessment, and Perspective (opens in a new tab)
- Findings of ACLProMediate: A Socio-cognitive Framework for Evaluating Proactive Agents in Multi-party Negotiation (opens in a new tab)
- EACLWhat's Missing in Vision-Language Models? Probing Their Struggles with Causal Order Reasoning (opens in a new tab)
- ACL (T1)WildFeedback: Aligning LLMs With In-situ User Interactions And Feedback (opens in a new tab)
202510 papers
- EMNLP (T1)AI Sees Your Location — But With a Bias Toward the Wealthy World (opens in a new tab)
- EMNLP (T1)Analyzing Uncertainty of LLM-as-a-Judge: Interval Evaluations with Conformal Prediction (opens in a new tab)
- Findings of EMNLPCan LLMs Express Personality Across Cultures? Introducing CulturalPersonas for Evaluating Trait Alignment (opens in a new tab)
- ACL (T1)Cross-lingual Pitfalls: Automatic Probing of Cross-lingual Weakness of Multilingual Large Language Models (opens in a new tab)
- COLMDiscovering Knowledge Deficiencies of Language Models on Massive Knowledge Base (opens in a new tab)
- AI4Research @ AAAIDrugAgent: Automating AI-Aided Drug Discovery Programming Through LLM Multi-Agent Collaboration (opens in a new tab)
- NLP4PIMultilingual Large Language Models Leak Human Stereotypes Across Language Boundaries (opens in a new tab)
- ICLR (T1)MUSE: Machine Unlearning Six-Way Evaluation for Language Models (opens in a new tab)
- Findings of EMNLPThe Hallucination Tax of Reinforcement Finetuning (opens in a new tab)
- EMNLP (T1)VISBIAS: Measuring Explicit and Implicit Social Biases in Vision Language Models (opens in a new tab)
202410 papers
- EMNLP (T1)"You Gotta be a Doctor, Lin": An Investigation of Name-Based Bias of Large Language Models in Employment Recommendations (opens in a new tab)
- ICML (T1)Adapting Static Fairness to Sequential Decision-Making: Bias Mitigation Strategies towards Equal Long-term Benefit Rate (opens in a new tab)
- NLP4PI @ EMNLPCLIMB: A Benchmark of Clinical Bias in Large Language Models (opens in a new tab)
- IEEE Data Engineering BulletinDoes Differential Privacy Impact Bias in Pretrained Language Models? (opens in a new tab)
- NAACL (T1)Fair Abstractive Summarization of Diverse Perspectives (opens in a new tab)
- EMNLP (T1)Images Speak Louder Than Words: Understanding and Mitigating Bias in Vision-Language Models from a Causal Mediation Perspective (opens in a new tab)
- EMNLP (T1)InterIntent: Investigating Social Intelligence of LLMs via Intention Understanding in an Interactive Game Context (opens in a new tab)
- NAACL · SeT-LLMSafer-Instruct: Aligning Language Models with Automated Preference Data (opens in a new tab)
- Findings of EMNLPSCORE: A Framework for Self-Contradictory Reasoning Evaluation (opens in a new tab)
- ICML (T1)TrustLLM: Trustworthiness in Large Language Models (opens in a new tab)
20237 papers
- EMNLP (T1)A Rose by Any Other Name Would Not Smell as Sweet: Social Bias in Name Mistranslations (opens in a new tab)
- Findings of EMNLPAre Personalized Stochastic Parrots More Dangerous? Evaluating Persona Biases in Dialogue Systems (opens in a new tab)
- W3PHIAIAuditing Algorithmic Fairness in Machine Learning for Health with Severity-based LOGAN (opens in a new tab)
- BEA @ ACLDoes BERT Exacerbate Gender or L1 Biases in Automated English Speaking Assessment? (opens in a new tab)
- WiNLPMind What You Measure For: A Study on Reliability of Prompt-Based Bias Measurement (opens in a new tab)
- EACLSODAPOP: Open-Ended Discovery of Social Biases in Social Commonsense Reasoning Models (opens in a new tab)
- NeurIPS (T1)TACO: Temporal Latent Action-Driven Contrastive Loss for Visual Reinforcement Learning (opens in a new tab)
Where alumni went19 listed, 40% to industry
Industry 4Academia 6
Most often to
- Amazon
- ByteDance
Also to
- Carnegie Mellon University
- UT Austin
Apply to USCFall 2027
Computer Science PhD (Viterbi)
- Round
- Single round (Dec 15)
- Opens
- —
- GRE
- See program page
- English
- See program page
- Fee
- See program page
Confirmed on the program page.
Similar labsnearest by research summary
- CNTsinghua
Ning Ding
- Large language models
- USUCLA
Kai-Wei Chang
- Large language models
- USMIT
Paul Liang
21 T1 - CNZhejiang
Zhixuan Chu
- Large language models
- GBImperial
Edoardo Maria Ponti
- Large language models
- USUC Berkeley
Dan Klein
- Multilingual, translation & linguistics
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.
USCcloses