Arnav Kumar Jain
Lead author of WEAVER, building a research program on learned world models for robot control at Mila.
“WEAVER achieves a Pearson correlation of ρ=0.870 between simulated and real-world success rates... finetuning on synthetic data closely matches that on real data, with only a 4% average performance gap”
Source→“WEAVER is about 20× faster than Ctrl-World inference pipeline on an RTX A6000 Ada GPU, and batched sampling scales sublinearly with the number of candidates, showing that our inference optimizations make world-model-based test-time planning practical for real-time manipulation”
Source→“we apply WEAVER in robotic hardware, demonstrating its effectiveness at policy improvement (real-world success rate improvement of 38% on top of the π0.5 robot foundation model)”
Source→“each view is encoded into H×W patch tokens using the pretrained Stable Diffusion 3 VAE encoder”
Source→AI-extracted from podcast / newsletter / paper summaries. May contain errors.