Repository associated with paper titled "CLP: Finetuning Vision-Language-Action Models Requires Fewer Layers Than You Think"

Repository associated with paper titled "RoboGaze: Evaluating Robot World Models via Structured Vision-Language Analysis"

Training non-differentiable networks via optimal transport.

🔥 A curated roadmap to the Efficient VLA landscape. We’re keeping this list live—contribute your latest work!

Repository associated with paper titled "TokenRatio: Principled Token-Level Preference Optimization via Ratio Matching", presented at ICML 2026.

Repository associated with paper titled "FOCA: Future-Oriented Conditioning for Data-Efficient Vision-Language-Action Adaptation", presented at ICML 2026.

A Curated List of Vision-Language-Action (VLA) and World Action Models (WAM) Research and Beyond

A unified inference runtime for VLA models.

Training-free structured sparsification for SAM.

[ICLR_2026] A Scalable Fragment-Aware Conformer Ensemble Transformer

OBEYED-VLA: Clutter-Resistant Vision-Language-Action Models through Object-Centric and Geometry Grounding

Official Implementation of "Interpretable 3D Neural Object Volumes for Robust Conceptual Reasoning." ICLR 2026.

[TMLR] DuFal: Dual-Frequency-Aware Learning for High-Fidelity Extremely Sparse-view CBCT Reconstruction

[ICLR 2024] MARCEL: Machine Learning over Molecular Conformer Ensembles

[NeurIPS 2025] How Many Tokens Do 3D Point Cloud Transformer Architectures Really Need?

S-Chain: Structured Visual Chain-of-Thought For Medicine

[ICML2025, NeurIPS2025 Spotlight] Sparse VideoGen 1 & 2: Accelerating Video Diffusion Transformers with Sparse Attention

[NeurIPS 2025] ExGra-Med: Medical Multi-Modal LLM with Extended Context Alignment

lecture slides for Deepmind x UCL 2021 reinforcement learning course available in YouTbue

Official implementation of Chain-of-Action: Trajectory Autoregressive Modeling for Robotic Manipulation. Accepted in NeurIPS 2025.

Hierarchical Refinement: Optimal Transport to Infinity and Beyond

JAX - A curated list of resources https://github.com/google/jax

Official implementation of "OneTwoVLA: A Unified Vision-Language-Action Model with Adaptive Reasoning"