Can VLA Models Learn from Real-World Data Continually without Forgetting?
A robotics research paper on Can VLA Models Learn from Real-World Data Continually without Forgetting?.
Engineering 5.0 · Research 7.0 · Business 4.0
Vision-Language-Action Models Papers, Code and Reading Roadmap
VLA models connect visual perception, language instructions and robot actions for embodied AI systems.
Technical roadmap
Read foundational VLA papers, then compare robot datasets, open-source implementations, action tokenization and real-world deployment limitations.
FineVLA: Fine-Grained Instruction Alignment for Steerable Vision-Language-Action Policies
A robotics research paper on FineVLA: Fine-Grained Instruction Alignment for Steerable Vision-Language-Action Policies.
Engineering 5.0 · Research 7.0 · Business 4.0
EXPO-FT: Sample-Efficient Reinforcement Learning Finetuning for Vision-Language-Action Models
A robotics research paper on EXPO-FT: Sample-Efficient Reinforcement Learning Finetuning for Vision-Language-Action Models.
Engineering 5.0 · Research 7.0 · Business 4.0
OASIS: Observation-Action Space Alignment via SE(3) Trajectory Prediction for Robotic Manipulation
A robotics research paper on OASIS: Observation-Action Space Alignment via SE(3) Trajectory Prediction for Robotic Manipulation.
Engineering 5.0 · Research 7.0 · Business 4.0
$π_0$-EqM: Equilibrium Matching for Closed-Loop Vision-Language-Action Control
A robotics research paper on $π_0$-EqM: Equilibrium Matching for Closed-Loop Vision-Language-Action Control.
Engineering 5.0 · Research 7.0 · Business 4.0
TactileReflex: Noise-Statistics-Driven Vision-Tactile Reflex Control for Force-Sensitive Manipulation
A robotics research paper on TactileReflex: Noise-Statistics-Driven Vision-Tactile Reflex Control for Force-Sensitive Manipulation.
Engineering 5.0 · Research 7.0 · Business 4.0
Afford-VLA: Action-Aligned Visual Planning via Internalized Affordance
A robotics research paper on Afford-VLA: Action-Aligned Visual Planning via Internalized Affordance.
Engineering 5.0 · Research 7.0 · Business 4.0
Agentic-VLA: Efficient Online Adaptation for Vision-Language-Action Models
A robotics research paper on Agentic-VLA: Efficient Online Adaptation for Vision-Language-Action Models.
Engineering 5.0 · Research 7.0 · Business 4.0
GesVLA: Gesture-Aware Vision-Language-Action Model Embedded Representations
A robotics research paper on GesVLA: Gesture-Aware Vision-Language-Action Model Embedded Representations.
Engineering 5.0 · Research 7.0 · Business 4.0
Lost in Fog: Sensor Perturbations Expose Reasoning Fragility in Driving VLAs
A robotics research paper on Lost in Fog: Sensor Perturbations Expose Reasoning Fragility in Driving VLAs.
Engineering 5.0 · Research 7.0 · Business 4.0
Beyond Binary Success: A Diagnostic Meta-Evaluation Framework for Fine-Grained Manipulation
A robotics research paper on Beyond Binary Success: A Diagnostic Meta-Evaluation Framework for Fine-Grained Manipulation.
Engineering 5.0 · Research 7.0 · Business 4.0
RIO: Flexible Real-Time Robot I/O for Cross-Embodiment Robot Learning
A robotics research paper on RIO: Flexible Real-Time Robot I/O for Cross-Embodiment Robot Learning.
Engineering 5.0 · Research 7.0 · Business 4.0
RankQ: Offline-to-Online Reinforcement Learning via Self-Supervised Action Ranking
A robotics research paper on RankQ: Offline-to-Online Reinforcement Learning via Self-Supervised Action Ranking.
Engineering 5.0 · Research 7.0 · Business 4.0
SABER: A Scalable Action-Based Embodied Dataset for Real-World VLA Adaptation
A robotics research paper on SABER: A Scalable Action-Based Embodied Dataset for Real-World VLA Adaptation.
Engineering 5.0 · Research 7.0 · Business 4.0
Toward Visually Realistic Simulation: A Benchmark for Evaluating Robot Manipulation in Simulation
A robotics research paper on Toward Visually Realistic Simulation: A Benchmark for Evaluating Robot Manipulation in Simulation.
Engineering 5.0 · Research 7.0 · Business 4.0
CoRAL: Contact-Rich Adaptive LLM-based Control for Robotic Manipulation
A robotics research paper on CoRAL: Contact-Rich Adaptive LLM-based Control for Robotic Manipulation.
Engineering 5.0 · Research 7.0 · Business 4.0