Afshin Dehghan

AToken: A Unified Tokenizer for Vision

Generation by Search: Scaling Test-Time Compute for Autoregressive Image Generation

Rooms from Motion: Un-posed Indoor 3D Object Detection as Localization and Mapping

MM1.5: Methods, Analysis & Insights from Multimodal LLM Fine-tuning

Understanding Alignment in Multimodal LLMs: A Comprehensive Study

FlexTok: Resampling Images into 1D Token Sequences of Flexible Length

StreamBridge: Turning Your Offline Video Large Language Model into a Proactive Streaming Assistant

SlowFast-LLaVA-1.5: A Family of Token-Efficient Video Large Language Models for Long-Form Video Understanding

UniGen: Enhanced Training & Test-Time Strategies for Unified Multimodal Understanding and Generation

SlowFast-LLaVA: A strong training-free baseline for video large language models