Trevor Darrell

Reconstruction Alignment Improves Unified Multimodal Models

Search Arena: Analyzing Search-Augmented LLMs

Person-Centric Annotations of LAION-400M: Auditing Bias and Its Transfer to Models

Constantly Improving Image Models Need Constantly Improving Benchmarks

DAVE: A VLM Vision Encoder for Document Understanding and Web Agents

Visualizing Thought: Conceptual Diagrams Enable Robust Planning in LMMs

Learning to Grasp Anything By Playing with Random Toys

Are Large Reasoning Models Interruptible?

LeVERB: Humanoid Whole-Body Control with Latent Vision-Language Instruction

Robotic Steering: Mechanistic Finetuning for Vision-Language-Action Models

FoundationMotion: Auto-labeling and Reasoning about Spatial Movement in Videos

Activation Reward Models for Few-Shot Model Alignment

Generate, but Verify: Reducing Hallucination in Vision-Language Models with Retrospective Resampling

Learning Adaptive Parallel Reasoning with Language Models

MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motion

Revealing and Reducing Gender Biases in Vision and Language Assistants (VLAs)

Whole-Body Conditioned Egocentric Video Prediction

Hidden in plain sight: VLMs overlook their visual representations

Video Action Differencing

REOrdering Patches Improves Vision Models

A Coefficient Makes SVRG Effective

Vision-Language Models Create Cross-Modal Task Representations

Visual Haystacks: A Vision-Centric Needle-In-A-Haystack Benchmark

SegLLM: Multi-round Reasoning Segmentation with Large Language Models

Analyzing the Language of Visual Tokens

Pre-training Auto-regressive Robotic Models with 4D Representations

VibeCheck: Discover and Quantify Qualitative Differences in Large Language Models

Pose Priors from Language Models

Data-Centric AI Governance: Addressing the Limitations of Model-Focused Policies

Wolf: Accurate Video Captioning with a World Summarization Framework

Task Vectors are Cross-Modal

Navigating the Labyrinth: Evaluating and Enhancing LLMs’ Ability to Reason About Search Problems