Trevor Darrell
Professor in Residence
Develops visual recognition algorithms and multimodal vision-language models.
Trevor Darrell's lab at UC Berkeley works on vision-language & multimodal, recognition, detection & segmentation and multimodal & grounded language. 114 papers since 2023, 92 at top venues, mostly CVPR and ICLR. 11 PhD students. 13 alumni. Applications close .
At a glance
- papers, 2023–2026
- 30/ year
- at T1 venues
- 81%
- PhD students
- 11
- MS students
- 0
- alumni listed
- 13
Papers by yearfrom DBLP, 2023–2026
- 14 papers2026 · T1 12 · collecting
- 34 papers2025 · T1 28
- 43 papers2024 · T1 33
- 23 papers2023 · T1 19
T1 venuesOtherVenues: CVPR, ICLR, NeurIPS, ICML, ECCV
Research focuspaper titles that use each topic's words
- Vision-language & multimodal45 papers
- Multimodal & grounded language33 papers
- Recognition, detection & segmentation10 papers
- vision-language models4 papers
- visual recognition0 papers
Paper list
From DBLP.
202614 papers
- Findings of ACLActivation Reward Models for Few-Shot Model Alignment (opens in a new tab)
- CVPR (T1)Attend Before Attention: Efficient and Scalable Video Understanding via Autoregressive Gazing (opens in a new tab)
- ECCV (T1)Chain-of-Visual-Thought: Teaching VLMs to See and Think Better with Continuous Visual Tokens (opens in a new tab)
- ICLR (T1)Constantly Improving Image Models Need Constantly Improving Benchmarks (opens in a new tab)
- ICLR (T1)DAVE: A VLM Vision Encoder for Document Understanding and Web Agents (opens in a new tab)
- CVPR (T1)It's Never Too Late: Noise Optimization for Collapse Recovery in Trained Diffusion Models (opens in a new tab)
- CVPR (T1)Latent Implicit Visual Reasoning (opens in a new tab)
- ICLR (T1)Learning to Grasp Anything By Playing with Random Toys (opens in a new tab)
- ECCV (T1)Lifting Embodied World Models for Planning and Control (opens in a new tab)
- Trans. Mach. Learn. Res.Navigating the Labyrinth: Evaluating LLMs' Ability to Reason About Search Problems (opens in a new tab)
- ICLR (T1)Person-Centric Annotations of LAION-400M: Auditing Bias and Its Transfer to Models (opens in a new tab)
- ICLR (T1)Reconstruction Alignment Improves Unified Multimodal Models (opens in a new tab)
202534 papers
- ICLR (T1)A Coefficient Makes SVRG Effective (opens in a new tab)
- CAIA Roadmap for Alignable Algorithmic Decision-Makers in the Medical Triage Domain (opens in a new tab)
- CVPR (T1)AutoPresent: Designing Structured Visuals from Scratch (opens in a new tab)
- ASRUCLAIRA: Leveraging Large Language Models to Judge Audio Captions (opens in a new tab)
- ICRA (T1)Decentralized Vehicle Coordination: The Berkeley DeepDrive Drone Dataset and Consensus-Based Models (opens in a new tab)
- ICCV (T1)Describe Anything: Detailed Localized Image and Video Captioning (opens in a new tab)
- ICCV (T1)Discovering Divergent Representations Between Text-To-Image Models (opens in a new tab)
- Findings of EMNLPDo What? Teaching Vision-Language-Action Models to Reject the Impossible (opens in a new tab)
- ICCV (T1)Dual-Process Image Generation (opens in a new tab)
- ICCV (T1)Enhancing Few-Shot Vision-Language Classification With Large Multimodal Model Features (opens in a new tab)
- ACL (T1)Enough Coin Flips Can Make LLMs Act Bayesian (opens in a new tab)
- NeurIPS (T1)Generate, but Verify: Reducing Hallucination in Vision-Language Models with Retrospective Resampling (opens in a new tab)
202443 papers
- Findings of ACLAligning Large Multimodal Models with Factually Augmented RLHF (opens in a new tab)
- NAACL (T1)ALOHa: A New Measure for Hallucination in Captioning Models (opens in a new tab)
- CVPR (T1)Compositional Chain-of-Thought Prompting for Large Multimodal Models (opens in a new tab)
- NeurIPS (T1)ConMe: Rethinking Evaluation of Compositional Reasoning for Modern VLMs (opens in a new tab)
- CVPR (T1)Describing Differences in Image Sets with Natural Language (opens in a new tab)
- ECCV (T1)EgoPet: Egomotion and Interaction Data from an Animal's Perspective (opens in a new tab)
- ECCV (T1)Finding Visual Task Vectors (opens in a new tab)
- CVPR (T1)From Audio to Photoreal Embodiment: Synthesizing Humans in Conversations (opens in a new tab)
- NeurIPS (T1)Humanoid Locomotion as Next Token Prediction (opens in a new tab)
- ICML (T1)Hyperbolic Active Learning for Semantic Segmentation under Domain Shift (opens in a new tab)
- Trans. Mach. Learn. Res.IMProv: Inpainting-based Multimodal Prompting for Computer Vision Tasks (opens in a new tab)
- ICLR (T1)Initializing Models with Larger Ones (opens in a new tab)
202323 papers
- CVPR (T1)Back to the Source: Diffusion-Driven Adaptation to Test-Time Corruption (opens in a new tab)
- ICCV (T1)Can Language Models Learn to Listen? (opens in a new tab)
- EMNLP (T1)CLAIR: Evaluating Image Captions with Large Language Models (opens in a new tab)
- NeurIPS (T1)Diffusion Hyperfeatures: Searching Through Time and Space for Semantic Correspondence (opens in a new tab)
- NeurIPS (T1)Diversify Your Vision Datasets with Automatic Diffusion-based Augmentation (opens in a new tab)
- ICML (T1)Dropout Reduces Underfitting (opens in a new tab)
- TPAMI (T1)Exploring Simple and Transferable Recognition-Aware Image Processing (opens in a new tab)
- EMNLP (T1)From Wrong To Right: A Recursive Approach Towards Vision-Language Explanation (opens in a new tab)
- ICML (T1)Guiding Pretraining in Reinforcement Learning with Large Language Models (opens in a new tab)
- NeurIPS (T1)Hierarchical Open-vocabulary Universal Image Segmentation (opens in a new tab)
- EMNLP (T1)Incorporating Structured Representations into Pretrained Vision & Language Models Using Scene Graphs (opens in a new tab)
- NeurIPS (T1)Large Language Models are Visual Reasoning Coordinators (opens in a new tab)
Where alumni went13 listed
Apply to UC BerkeleyFall 2027
UC Berkeley EECS Computer Science PhD
- Round
- Single round
- Opens
- —
- GRE
- not required
- English
- TOEFL/IELTS (non-English-country degrees)
- Fee
- See program page
Confirmed on the program page.
Similar labsnearest by research summary
- USUSC
C.-C. Jay Kuo
- Video understanding
- CHEPFL
Andrea Cavallaro
- Recognition, detection & segmentation
- CNTsinghua
Liangrui Peng
- Interfaces & interaction techniques
- USUMD
Soheil Feizi
- Trustworthy & safe ML
- GBOxford
Philip Torr
- Trustworthy & safe ML
- USUMD
Tom Goldstein
- Trustworthy & safe ML
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- DBLP
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
UC Berkeleycloses