New paper shows DiT latents can be linearly probed to predict pixels, skipping the VAE decoder
kwangmoo_yi · x · 2026-11-19
Chen et al.'s paper "Decoding Without a Decoder: Direct Pixel Prediction from Diffusion Transformers" shows you can linearly probe the DiT latent space to skip the decoder entirely. The poster raises the provocative question: was training with a V(Q)AE just a speed optimization that can now be thrown away?
Related event: DWD Paper: Diffusion Models Predict Pixels Without a Decoder(4 posts)→
More from Research
- New preprint asks: if LLM interpretability succeeded completely, what would it teach us about human language? — najoungkim · 2026-11-19
- Researcher: my real peer review is asking the world's top 2-3 experts, conference reviews are noise — taiyasaki · 2026-11-19
- A policymaker's primer on AI benchmarks: five questions to ask before trusting scores — sethlazar · 2026-11-19
- WeightBench: New Benchmark Tests LLMs as Weight-Space Programmers — Aayush Marasini · 2026-11-19
- Study of 40 LLM Tech Reports: Only 19 of 73 Novelty Claims Have Ablations — Bahaeddin Türkoğlu · 2026-11-19
- GPT-5.6 Luna reranking barely improves from top-20 to top-100 docs at 5x the cost — eshorten300 · 2026-11-19