
RLE at MIT
1.9K posts

RLE at MIT
@RLEatMIT
The Research Laboratory of Electronics (RLE) at the Massachusetts Institute of Technology (MIT)



















🎥 Video generation is hitting the memory wall. As videos get longer, the KV cache quietly explodes — and long-horizon consistency starts to break. We built Quant VideoGen: a training-free KV cache compression method for auto-regressive video diffusion. Instead of storing every KV in high precision, QVG exploits video’s spatiotemporal redundancy with semantic-aware smoothing + progressive residual quantization. 🚀 Up to 7× KV memory reduction ⚡ <4% overhead ✅ Strong long-video quality 🕹️ Deploy HYWorldPlay on your own RTX 5090 locally KV compression is becoming a core scaling primitive — not just for LLMs, but for video generation too. Paper: arxiv.org/abs/2602.02958 Code: github.com/svg-project/Qu… (1/5)














