MINT-SJTU: Machine Intelligence and Interaction Lab @ SJTU
Repositories
MINT-SJTU/Evo-RLT
MINT-SJTU/RoboClaw
RoboClaw is an Embodied AI Assistant.
MINT-SJTU/RoboFAC.io
MINT-SJTU/Evo-SOTA.io
This website is for the collection of VLA SOTA results.
MINT-SJTU/Evo-Depth
MINT-SJTU/LeRobot-Anything-U-Arm
U-Arm: Lerobot-Everything-Cross-Embodiment-Teleoperation
MINT-SJTU/Evo-RL
We release Evo-RL, the opensource real-world offline RL on So-101 and AgileX PiPER for easier reproduction.
MINT-SJTU/Evo-1
Evo-1: Lightweight Vision-Language-Action Model with Preserved Semantic Alignment
MINT-SJTU/VLA-Pruner
The official implementation of VLA-Pruner: Temporal-Aware Dual-Level Visual Token Pruning for Efficient Vision-Language-Action Inference.
MINT-SJTU/LA4VLA
MINT-SJTU/RoboFAC
RoboFAC: A Comprehensive Framework for Robotic Failure Analysis and Correction
MINT-SJTU/EgoGrasp
MINT-SJTU/STI-Bench
STI-Bench : Are MLLMs Ready for Precise Spatial-Temporal World Understanding?
MINT-SJTU/SIM2REAL2SIM
MINT-SJTU/Evo-0
Evo-0: Vision-Language-Action Model with Implicit Spatial Understanding.
MINT-SJTU/mint-sjtu.github.io
MINT-SJTU/Evo-data.io
MINT-SJTU/EgoGrasp.io
MINT-SJTU/Evo-temp.io
MINT-SJTU/Evo-1.io
MINT-SJTU/Evo-0.io
The project page of Evo-0