Stars
Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models. TMLR 2025.
Index of hardware design repositories — CPUs, arithmetic units, SoC design, HDL, and power electronics
🌐 Vision-Language-Action Models for Autonomous Driving: Past, Present, and Future
[AAAI 2026] OpenDriveVLA: Towards End-to-end Autonomous Driving with Large Vision Language Action Model
[CVPR 2026] Visual Geometry Transformer for Autonomous Driving
[CVPR'26] QuantVLA: Scale-Calibrated Post-Training Quantization for Vision-Language-Action Models
InternVLA-A1: Unifying Understanding, Generation, and Action for Robotic Manipulation
UniDriveVLA: Unifying Understanding, Perception, and Action Planning for Autonomous Driving
RTL design and verification for LLM hardware accelerators — systolic arrays, attention engines, dataflow architectures, and FPGA/ASIC implementation
Vega: Learning to Drive with Natural Language Instructions
A high-performance, cycle-accurate simulation environment for a custom neural network accelerator architecture. It provides a complete toolchain that bridges the gap between high-level neural netwo…
🔥 A curated roadmap to the Efficient VLA landscape. We’re keeping this list live—contribute your latest work!
A method to increase the speed and lower the memory footprint of existing vision transformers.
DRAMsim3: a Cycle-accurate, Thermal-Capable DRAM Simulator
opensource NPU for LLM inference (this run gpt2)
A node-based visual tool for organizing thoughts and notes in a non-linear way.
[IEEE TIV] OccFusion: Multi-Sensor Fusion Framework for 3D Semantic Occupancy Prediction
[NeurIPS2025 Spotlight] Implementation of "GaussianFusion: Gaussian-Based Multi-Sensor Fusion for End-to-End Autonomous Driving"
Evo-1: Lightweight Vision-Language-Action Model with Preserved Semantic Alignment
[ICCV 2025] A Simple yet Effective Pathway to Empowering LLaVA to Understand and Interact with 3D World
[CVPR 2023] An academic alternative to Tesla's occupancy network for autonomous driving.
[ECCV 2022] This is the official implementation of BEVFormer, a camera-only framework for autonomous driving perception, e.g., 3D object detection and semantic map segmentation.
Repository to host and maintain SCALE-Sim code
BEVFormer inference on TensorRT, including INT8 Quantization and Custom TensorRT Plugins (float/half/half2/int8).
[DATE'2025, TCAD'2025] Terafly : A Multi-Node FPGA Based Accelerator Design for Efficient Cooperative Inference in LLMs
[Lumina具身智能社区] 具身智能技术指南 Embodied-AI-Guide
[IV'24] UniBEV: the official implementation of UniBEV