★ News ★

Oct'26:

• Invited talk at George Mason University

Sep'26:

• 3 papers accepted to NeurIPS 2026!

  • Aligning Language Model Benchmarks with Pairwise Preferences

  • AI Construct Lexis: An Ontology of the Hidden Assumptions in AI Evaluation

  • MedKIT: Evaluating Knowledge Integration and Generalization in Large Language Models

• Received Hewlett Foundation Award on methods to bridge the benchmark-to-reality gap for AI in biosecurity ($3.4M)

• Serving as AC for ICLR'27

Aug'26:

• NSF award on Research on Research Security

• Invited talk at Google Research on Building AI that Reasons About Time Series

June'26:

• Facilitated 2026 AHLI Summer Camp and organized CHIL 2026 in Seattle.

May'26:

• I have received an Apple Faculty Research Award

• First two RAVE group PhD students graduated. Each won their respective school's PhD dissertation award! 🎉

  • Bryan Christ (Data Science PhD; 2026 UVA SDS Distinguished Dissertation Award)

  • Matt Landers (CS PhD; 2026 UVA SEAS Outstanding PhD Award)

• Alex Gates, Steven Johnson, and I were awarded an AI & Democracy Seed Grant

• Paper accepted to Obesity Surgery on using retrieval-augmented generation to educate patients about bariatric surgery

• Workshop accepted to COLM 2026 on LLM/VLM Deployment Opportunities and Risks in Healthcare

• Invited panelist for UVA's Environmental Institute's Environment + AI Workshop

• New preprint on Living Benchmarks for Event-Augmented Time Series Forecasting

• Serving as AC for NeurIPS'26

Apr'26:

• Three papers accepted to ICML 2026

  • BEDTime: A Unified Benchmark for Automatically Describing Time Series

  • ReasonEdit: Editing Vision-Language Models using Human Reasoning

  • LLMs as Noisy Channels: A Shannon Perspective on Model Capacity and Scaling Laws

• Co-Investigator on NIH R01 as on modeling long-term trajectories in diabetes

• Keynote address at UVA's Research Computing Exhibition

• Four papers accepted to ACL 2026

  • Main Track

    • Inferring Events from Time Series Using Language Models

    • EDUMATH: Generating Standards-aligned Educational Math Word Problems

    • Evaluating Temporal Consistency in Multi-Turn Language Models

  • Findings Track

    • A Survey of Toxicity Mitigation Strategies for Multilingual Language Models

• Paper accepted to CHIL 2026 on using language models to identify side effects of breast cancer treatments

Mar'26:

• Congratulations to group member Matt Landers for defending his PhD on Reinforcement Learning in Combinatorial Action Spaces!

• I have passed my mid-tenure renewal... time flies!

• 3Cavaliers Funding Award  ($80k) as PI to work on Automated Post-Surgery Feedback: AI-Driven Time Series Analysis to Improve Anesthesia Care.

• Invited talk in UVA's Center for Advanced Medical Analytics

Feb'26:

• Invited talk at UMD on Continually-Editing VLMs

• New preprints on aligning benchmarks with human preferences and editing VLMs using human reasoning.

Jan'26:

• Two papers accepted to ICLR 2026 on accelerating offline RL and uncertainty quantification for LLMs.

• Paper accepted to MLSys on accelerating diffusion-based language models

• Paper accepted to JAMIA on extracting social determinants of health from clinical notes

Dec'25:

• Best paper award at NeurIPS 2025 AI4D3 Workshop for Constrained Molecular Generation with Discrete Diffusion for Drug Discovery

Nov'25:

• Paper accepted to KDD 2026 on Instruction-Based Time Series Editing. Congratulations, Joy!

• Featured interview on the AI Exchange @ UVA Podcast

• Paper accepted to TMLR on Adaptively mixing language model training data via bayesian optimization

Oct'25:

• Congratulations to group member Bryan Christ for defending his PhD!

• Paper accepted to Scientific Reports on subtyping for Type 2 Diabetes

Sept'25:

• Awarded $2M NIH grant over 2 years to develop A Model Editing Framework for Participatory Multimodal AI in Dermatology (with Ahmed Alaa and Roxana Daneshjou)

• 3 papers accepted to NeurIPS'25!

• Paper accepted to NPJ Digital Medicine paper on risks of medical misinformation from LLMs

• New preprints:

  • Benchmark for automatically describing time series

  • Uncertainty quantification for language models

Aug'25:

• 4 papers accepted to EMNLP'25!

  • Extracting Hidden Moderation Criteria from Reddit Communities

  • Sparse Autoencoder Features for Classifications and Transferability

  • ModelCitizens: Representing Community Voices in Online Safety

  • Lifelong Knowledge Editing requires Better Regularization

• Paper accepted to CIKM'25!

  • Backdoor in Seconds: Unlocking Vulnerabilities in Large Pre-trained Models via Model Editing

July'25:

• Paper accepted to COLM'25!

  • PolyGuard: A Multilingual Safety Moderation Tool for 17 Languages

• Organizing NeurIPS'25 Workshop on Learning from Time Series for Healthcare

June'25:

• Paper accepted to Rehabilitation Psychology on biased portrayals of disability by VLMs

• Featured in faculty research spotlight

May'25:

• 4 papers accepted to ACL'25!

  • Scaling laws for LLM quantization

  • Isolating and editing math skills in LLMs

  • Efficient knowledge editing

  • Model editing with external graph-based memory

• 3 papers accepted to ICML'25!

  • WikiBigEdit: Understanding the limits of lifelong knowlege editing

  • BalancEdit: Dynamically Balancing the Generality-Locality Trade-off in Multi-modal Model Editing

  • Medical Large Language Model Benchmarks Should Prioritize Construct Validity (Oral)

March'25:

• Our AAAI'25 KnowFM Workshop paper was awarded the Outstanding Paper Award.

• New preprints:

  • Inferring event descriptions from time series with language models

  • Dialect bias in LLM reasoning

  • Constrained diffusion models

  • Survey on multi-modal time series models

Feb'25:

• New preprints on empirical investigations of sparse autoencoders and lifelong model editing

• Paper accepted to CPAL'25 on Sparse MoE

Jan'25:

• I have been awarded a CapitalOne Faculty Fellowship to work on Time Series Reasoning!

• 2 papers accepted to ICLR'25!

  • Composable Interventions for Language Models - Congrats, Arinbjorn!

  • Learning from Time Series under Label Noise

• Paper on sequential knowledge editing accepted to Workshop on Knowledgeable Foundation Models at AAAI 25

• New preprint on lifelong model editing

Dec'24:

• Invited talks at UVA's Genome Sciences Seminar Series and UVA's Darden Business School 

• New preprint on foundation models for protein phenotypes 

Nov'24:

• Lab member Xu Ouyang was awarded an iPRIME PhD Fellowship --- congrats Xu!

• New preprint on scaling laws for LLM quantization

Oct'24:

• New preprints:

  • attacking VLMs with model edits

  • LLMs resisting misinformation requests

  • high-dimensional action spaces in Offline RL. 

• Paper accepted to IEEE BigData on spike train classification

Sep'24:

• 3 papers accepted to NeurIPS'24!

  • Are Language Models Actually Useful for Time Series Forecasting? (Spotlight!) - Congrats, Mingtian!

  • Test-Time Debiasing of Vision-Language Embeddings

  • UniTS:  A Unified Multi-Task Time Series Model

• 3 papers accepted to EMNLP'24!

• MATHWELL: Generating Educational Math Word Problems with Teacher Annotations - Congrats, Bryan!

• Language Models Still Struggle to Zero-shot Reason about Time Series

• Language Models are Surprisingly Fragile to Drug Names in Biomedical Benchmarks

Aug'23:

• Paper accepted to TMLR on using LLMs for robust text classification

July'24:

• Paper accepted to COLM'24 on multilingual toxicity in LLMs.

• Paper accepted to AIES'24 on detecting implicit social biases in VL models

July'24:

• New preprints:

  • composable interventions for LLMs

  • extracting social determinants of health with LLMs

• Paper accepted to MICCAI'24 on federated learning for medical imaging

May'24:

• Paper accepted to ACL'24 on categorical knowledge editing for LLMs

Apr'24:

• Nature Medicine paper on bias in computational pathology

Spring'24: Invited talks at Dartmouth, IBM Research, UCSF/UC Berkeley, and the University of Alabama, Birmingham