Tuna-2: Pixel Embeddings Beat Vision Encoders for Multimodal Understanding and Generation
From Daily Paper Cast by Jingwen Liang, Gengyu Wang
April 29, 2026 · 22 min · Episode 1810
About this episode
The episode discusses Tuna-2, a unified multimodal model that utilizes pixel embeddings for visual understanding and generation, outperforming traditional vision encoders.
🤗 Upvotes: 47 | cs.CV Authors: Zhiheng Liu, Weiming Ren, Xiaoke Huang, Shoufa Chen, Tianhong Li, Mengzhao Chen, Yatai Ji, Sen He, Jonas Schult, Belinda Zeng, Tao Xiang, Wenhu Chen, Ping Luo, Luke Zettlemoyer, Yuren Cong Title: Tuna-2: Pixel Embeddings Beat Vision Encoders for Multimodal Understanding and Generation Arxiv: http://arxiv.org/abs/2604.24763v1 Abstract: Unified multimodal models typically rely on pretrained vision encoders and use separate visual representations for understanding and generation, creating misalignment between the two tasks and preventing fully end-to-end optimization from raw pixels. We introduce Tuna-2, a native unified multimodal model that performs visual understanding and generation directly based on pixel embeddings. Tuna-2 drastically simplifies the model architecture by employing simple patch embedding layers to encode visual input, completely discarding the modular vision encoder designs such as the VAE or the representation encoder. Experiments show that Tuna-2 achieves state-of-the-art performance in multimodal benchmarks, demonstrating that unified pixel-space modelling can fully compete with latent-space approaches for high-quality image…
People in this episode
Hosts: Jingwen Liang, Gengyu Wang
Topics covered
- multimodal models
- pixel embeddings
- visual understanding
- image generation
- model architecture
- state-of-the-art performance
Keywords
- Tuna-2
- pixel embeddings
- vision encoders
- multimodal understanding
- image generation
- model architecture
- state-of-the-art
- fine-grained visual perception
More episodes of Daily Paper Cast
- EvoArena: Tracking Memory Evolution for Robust LLM Agents in Dynamic Environments · June 13, 2026 · 24 min
- MiniMax Sparse Attention · June 13, 2026 · 26 min
- SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning · June 13, 2026 · 23 min
- InterleaveThinker: Reinforcing Agentic Interleaved Generation · June 13, 2026 · 21 min
- FORT-Searcher: Synthesizing Shortcut-Resistant Search Tasks for Training Deep Search Agents · June 13, 2026 · 23 min
- Robust-U1: Can MLLMs Self-Recover Corrupted Visual Content for Robust Understanding? · June 13, 2026 · 20 min
Explore listener stats, chart rankings, contacts and more on the Daily Paper Cast podcast page.