Skip to content
The Chat Archive
sonar-reasoning-pro-high LMArena Search Arena 24K

Generate my CV to include my publications. My name is Lisa Dunlap and I'm a researcher…

  1. User

    Generate my CV to include my publications. My name is Lisa Dunlap and I'm a researcher from UC Berkeley

  2. Assistant

    Lisa Dunlap is a PhD candidate in Computer Science at the University of California, Berkeley, affiliated with BAIR and RISELab, advised by Joseph E. Gonzalez, Trevor Darrell, and Sarah Bargal. Her research focuses on explainable AI, vision-language models, and scalable evaluation methods for machine learning systems. Below is a curated CV highlighting her publications:


    Education

    PhD in Computer Science
    University of California, Berkeley (2021–Present)
    Advisors: Joseph E. Gonzalez, Trevor Darrell, Sarah Bargal

    BA in Mathematics & Computer Science
    University of California, Berkeley (2017–2020)
    GPA: 3.75/4.0


    Research Experience

    PhD Researcher, BAIR/RISELab, UC Berkeley (2021–Present)

    • Developing evaluation frameworks for large language/vision models
    • Designing explainable AI systems and dataset augmentation techniques

    Machine Learning Researcher, Splunk (2020–2021)

    • Built robust NLP models with explainability features

    Selected Publications

    2024

    • From Crowdsourced Data to High-Quality Benchmarks: Arena-Hard and BenchBuilder Pipeline
      Tianle Li, Wei-Lin Chiang, Lisa Dunlap, et al.
      arXiv preprint arXiv:2406.11939 [160 citations*]

    • Describing Differences in Image Sets with Natural Language
      Lisa Dunlap, Yuxiao Zhang, Xin Wang, et al.
      CVPR 2024 [32 citations]

    2023

    • Diffusion Hyperfeatures: Searching Through Time and Space for Semantic Correspondence
      Grace Luo, Lisa Dunlap, Dong Huk Park, et al.
      NeurIPS 2023 [118 citations]

    • Diversify Your Vision Datasets with Automatic Diffusion-Based Augmentation
      Lisa Dunlap, Alyssa Umino, Han Zhang, et al.
      NeurIPS 2023 [65 citations]

    2020–2022

    • NBDT: Neural-Backed Decision Trees
      Alvin Wan, Lisa Dunlap, Daniel Ho, et al.
      ICLR 2021 [215 citations]

    • HyperSched: Dynamic Resource Reallocation for Model Development on a Deadline
      Richard Liaw, Romil Bhardwaj, Lisa Dunlap, et al.
      ACM SOCC 2019 [64 citations]


    Research Interests

    • Explainable/Advisable AI
    • Vision-Language Model Evaluation
    • Automated Dataset Curation
    • Resource-Efficient ML Systems

    Contact

    Email: [EMAIL_ADDRESS]
    Google Scholar: Lisa Dunlap[1]
    CV/Website: lisabdunlap.com

    *Citations updated as of April 2025. Preprints marked with asterisk.

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.