:round_pushpin: Session Details:
Date/Time: Wednesday, June 3 | Morning Session
Location: Poster I.199, Hall C
W/ Özgür Aslan and Cyrus Neary
Check out the project here: akuramshin.github.io/tread/
#ICRA2026 #Robotics #AI #MachineLearning #VisionLanguageAction
Date/Time: Wednesday, June 3 | Morning Session
Location: Poster I.199, Hall C
W/ Özgür Aslan and Cyrus Neary
Check out the project here: akuramshin.github.io/tread/
#ICRA2026 #Robotics #AI #MachineLearning #VisionLanguageAction
Task Robustness via Re-Labelling Vision-Action Robot Data
Task Robustness via Re-Labelling Vision-Action Robot Data
akuramshin.github.io
June 3, 2026 at 12:27 PM
:round_pushpin: Session Details:
Date/Time: Wednesday, June 3 | Morning Session
Location: Poster I.199, Hall C
W/ Özgür Aslan and Cyrus Neary
Check out the project here: akuramshin.github.io/tread/
#ICRA2026 #Robotics #AI #MachineLearning #VisionLanguageAction
Date/Time: Wednesday, June 3 | Morning Session
Location: Poster I.199, Hall C
W/ Özgür Aslan and Cyrus Neary
Check out the project here: akuramshin.github.io/tread/
#ICRA2026 #Robotics #AI #MachineLearning #VisionLanguageAction
Tanzende Roboter in China, aber leere Hallen in Europa? #Robotik-Chef Björn Enders verrät in #Dresden, was Robotern wirklich noch fehlt. #BjörnEnders #Symbiosika #HumanoideRoboter #PhysicalAI #VisionLanguageAction #KIRobotik #BrainsonSilicon
www.diesachsen.de
www.diesachsen.de
www.diesachsen.de
Schluss mit dem Hype: Wann humanoide Roboter wirklich arbeiten
www.diesachsen.de
September 16, 2026 at 11:06 AM
Tanzende Roboter in China, aber leere Hallen in Europa? #Robotik-Chef Björn Enders verrät in #Dresden, was Robotern wirklich noch fehlt. #BjörnEnders #Symbiosika #HumanoideRoboter #PhysicalAI #VisionLanguageAction #KIRobotik #BrainsonSilicon
www.diesachsen.de
www.diesachsen.de
Dancing robots in China, but empty halls in Europe? Robotics chief Björn Enders reveals in #Dresden what robots are still missing. #BjörnEnders #Symbiosika #HumanoidRobots #PhysicalAI #VisionLanguageAction #RoboticsinGermany #AIRobotics #BrainsonSilicon
www.diesachsen.de
www.diesachsen.de
www.diesachsen.de
Enough with the Hype: When Will Humanoid Robots Actually Be Working?
www.diesachsen.de
September 16, 2026 at 11:08 AM
Dancing robots in China, but empty halls in Europe? Robotics chief Björn Enders reveals in #Dresden what robots are still missing. #BjörnEnders #Symbiosika #HumanoidRobots #PhysicalAI #VisionLanguageAction #RoboticsinGermany #AIRobotics #BrainsonSilicon
www.diesachsen.de
www.diesachsen.de
Im Werk Landshut entwickelt BMW das Software-Fundament für humanoide Roboter. Statt starren Code nutzt man KI-Modelle und offene Plattformen. www.cio.de/article/4199... #physicalAI #bmw #AI #VisionLanguageAction #automobilindustrie
Physical AI: BMW baut offenes Roboter-Ökosystem
Im Werk Landshut entwickelt BMW das Software-Fundament für humanoide Roboter. Statt starren Code nutzt man KI-Modelle und offene Plattformen.
www.cio.de
July 21, 2026 at 9:24 AM
Im Werk Landshut entwickelt BMW das Software-Fundament für humanoide Roboter. Statt starren Code nutzt man KI-Modelle und offene Plattformen. www.cio.de/article/4199... #physicalAI #bmw #AI #VisionLanguageAction #automobilindustrie
Google DeepMind's Gemini Robotics On-Device is here!
This #VisionLanguageAction foundation model operates locally on robot hardware, enabling low-latency inference and can be fine-tuned for specific tasks with as few as 50 demonstrations.
👉 bit.ly/4ob3mQf
#Robotics #AI #GoogleDeepMind #InfoQ
This #VisionLanguageAction foundation model operates locally on robot hardware, enabling low-latency inference and can be fine-tuned for specific tasks with as few as 50 demonstrations.
👉 bit.ly/4ob3mQf
#Robotics #AI #GoogleDeepMind #InfoQ
July 17, 2025 at 7:37 AM
Google DeepMind's Gemini Robotics On-Device is here!
This #VisionLanguageAction foundation model operates locally on robot hardware, enabling low-latency inference and can be fine-tuned for specific tasks with as few as 50 demonstrations.
👉 bit.ly/4ob3mQf
#Robotics #AI #GoogleDeepMind #InfoQ
This #VisionLanguageAction foundation model operates locally on robot hardware, enabling low-latency inference and can be fine-tuned for specific tasks with as few as 50 demonstrations.
👉 bit.ly/4ob3mQf
#Robotics #AI #GoogleDeepMind #InfoQ
HyperVLA reduces inference load by activating only about 1% of VLA model parameters, delivering roughly a 120× speed boost while keeping zero‑shot success rates comparable. Read more: https://getnews.me/hypervla-cuts-vision-language-action-model-inference-cost-by-90x/ #hypervla #visionlanguageaction
October 8, 2025 at 5:36 AM
HyperVLA reduces inference load by activating only about 1% of VLA model parameters, delivering roughly a 120× speed boost while keeping zero‑shot success rates comparable. Read more: https://getnews.me/hypervla-cuts-vision-language-action-model-inference-cost-by-90x/ #hypervla #visionlanguageaction
SITCOM adds a learned dynamics model to Vision‑Language‑Action robots, raising task success from ~50% to ~75% in SIMPLER tests; it was trained on the BridgeV2 dataset. Read more: https://getnews.me/sitcom-boosts-long-horizon-planning-for-vision-language-action-robots/ #sitcom #visionlanguageaction
October 7, 2025 at 9:53 PM
SITCOM adds a learned dynamics model to Vision‑Language‑Action robots, raising task success from ~50% to ~75% in SIMPLER tests; it was trained on the BridgeV2 dataset. Read more: https://getnews.me/sitcom-boosts-long-horizon-planning-for-vision-language-action-robots/ #sitcom #visionlanguageaction
CogVLA reaches 97.4% success on LIBERO and cuts training costs by ~2.5×, while reducing inference latency by ~2.8×. The code and model weights are open‑sourced on GitHub. Read more: https://getnews.me/cogvla-boosts-vision-language-action-efficiency-via-routing/ #cogvla #visionlanguageaction
October 3, 2025 at 9:33 AM
CogVLA reaches 97.4% success on LIBERO and cuts training costs by ~2.5×, while reducing inference latency by ~2.8×. The code and model weights are open‑sourced on GitHub. Read more: https://getnews.me/cogvla-boosts-vision-language-action-efficiency-via-routing/ #cogvla #visionlanguageaction
Hybrid Training lets VLA models learn chain‑of‑thought but skip the thought step at inference, reducing token output. It retained performance pick‑and‑place tests. Read more: https://getnews.me/hybrid-training-cuts-cot-overhead-for-vision-language-action-models/ #visionlanguageaction #hybridtraining
October 2, 2025 at 11:48 PM
Hybrid Training lets VLA models learn chain‑of‑thought but skip the thought step at inference, reducing token output. It retained performance pick‑and‑place tests. Read more: https://getnews.me/hybrid-training-cuts-cot-overhead-for-vision-language-action-models/ #visionlanguageaction #hybridtraining
VLA‑RFT reaches robust performance with under 400 fine‑tuning steps, beating supervised baselines, and uses a data‑driven world model as a controllable simulator (Oct 2025). https://getnews.me/vision-language-action-reinforcement-fine-tuning-improves-robustness/ #visionlanguageaction #worldmodel
October 2, 2025 at 9:56 PM
VLA‑RFT reaches robust performance with under 400 fine‑tuning steps, beating supervised baselines, and uses a data‑driven world model as a controllable simulator (Oct 2025). https://getnews.me/vision-language-action-reinforcement-fine-tuning-improves-robustness/ #visionlanguageaction #worldmodel
World‑Env provides a simulator that lets Vision‑Language‑Action models continue safe RL training, achieving gains with as few as five expert demonstrations per task. Read more: https://getnews.me/world-env-safe-rl-post-training-for-vision-language-action-models/ #visionlanguageaction #worldenv
October 1, 2025 at 1:29 AM
World‑Env provides a simulator that lets Vision‑Language‑Action models continue safe RL training, achieving gains with as few as five expert demonstrations per task. Read more: https://getnews.me/world-env-safe-rl-post-training-for-vision-language-action-models/ #visionlanguageaction #worldenv
IA-VLA adds a vision-language model to enrich instructions, boosting success on tasks with identical objects. It outperformed baseline in duplicate-object tests. Read more: https://getnews.me/ia-vla-boosts-vision-language-action-models-for-complex-robot-tasks/ #visionlanguageaction #robotics
September 30, 2025 at 11:46 PM
IA-VLA adds a vision-language model to enrich instructions, boosting success on tasks with identical objects. It outperformed baseline in duplicate-object tests. Read more: https://getnews.me/ia-vla-boosts-vision-language-action-models-for-complex-robot-tasks/ #visionlanguageaction #robotics
Researchers show FreezeVLA can freeze Vision‑Language‑Action robots with a single adversarial image, achieving a 76.2% success rate across three leading VLA models. Read more: https://getnews.me/freezevla-action-freezing-attacks-on-vision-language-action-models/ #visionlanguageaction #adversarial
September 26, 2025 at 5:59 PM
Researchers show FreezeVLA can freeze Vision‑Language‑Action robots with a single adversarial image, achieving a 76.2% success rate across three leading VLA models. Read more: https://getnews.me/freezevla-action-freezing-attacks-on-vision-language-action-models/ #visionlanguageaction #adversarial
ThinkAct, presented at NeurIPS 2025, uses a dual‑system where an LLM plans and a visual latent vector guides an action model, improving long‑horizon planning and self‑correction. https://getnews.me/thinkact-visual-latent-planning-for-vision-language-action-ai/ #thinkact #visionlanguageaction
September 20, 2025 at 12:10 PM
ThinkAct, presented at NeurIPS 2025, uses a dual‑system where an LLM plans and a visual latent vector guides an action model, improving long‑horizon planning and self‑correction. https://getnews.me/thinkact-visual-latent-planning-for-vision-language-action-ai/ #thinkact #visionlanguageaction
NVIDIA Defines World-Action Models Era for Robotics with AI
#RoboticsAi #VisionLanguageAction #WorldModels
#RoboticsAi #VisionLanguageAction #WorldModels
June 15, 2026 at 12:35 PM
NVIDIA Defines World-Action Models Era for Robotics with AI
#RoboticsAi #VisionLanguageAction #WorldModels
#RoboticsAi #VisionLanguageAction #WorldModels
#Emergentcapabilities in #largelanguagemodels, such as in-context learning, can also appear in #visionlanguageaction (#VLA) models. Scaling up #roboticfoundationmodels allows for emergent human-to-robot transfer, improving performance on tasks demonstrated in human videos by approximately 2x.…
December 20, 2025 at 2:39 PM
#Emergentcapabilities in #largelanguagemodels, such as in-context learning, can also appear in #visionlanguageaction (#VLA) models. Scaling up #roboticfoundationmodels allows for emergent human-to-robot transfer, improving performance on tasks demonstrated in human videos by approximately 2x.…
Helix Revolutionizes Home Robotics with Cutting-Edge Vision-Language-Action Model
#HomeRobotics #VisionLanguageAction #HelixRobot
#HomeRobotics #VisionLanguageAction #HelixRobot
February 22, 2025 at 10:36 AM
Helix Revolutionizes Home Robotics with Cutting-Edge Vision-Language-Action Model
#HomeRobotics #VisionLanguageAction #HelixRobot
#HomeRobotics #VisionLanguageAction #HelixRobot
USIM and U0: A Vision-Language-Action Dataset and Model for General
Underwater Robots
Jian Wang, Junwen Gu et al.
Paper
Details
#USIMDataset #UnderwaterRobotics #VisionLanguageAction
Underwater Robots
Jian Wang, Junwen Gu et al.
Paper
Details
#USIMDataset #UnderwaterRobotics #VisionLanguageAction
October 12, 2025 at 4:01 PM
USIM and U0: A Vision-Language-Action Dataset and Model for General
Underwater Robots
Jian Wang, Junwen Gu et al.
Paper
Details
#USIMDataset #UnderwaterRobotics #VisionLanguageAction
Underwater Robots
Jian Wang, Junwen Gu et al.
Paper
Details
#USIMDataset #UnderwaterRobotics #VisionLanguageAction
Humanoid Robotics and NVIDIA Partner to Accelerate Humanoid Robot Development
Read more:
https://quantumzeitgeist.com/humanoid-robotics-and-nvidia-partner-to-accelerate-humanoid-robot-development/
Read more:
https://quantumzeitgeist.com/humanoid-robotics-and-nvidia-partner-to-accelerate-humanoid-robot-development/
Humanoid Robotics And NVIDIA Partner To Accelerate Humanoid Robot Development
Humanoid collaborates with NVIDIA to advance humanoid robot development utilising accelerated computing and simulation This partnership focuses on rapid prototyping via NVIDIA’s Isaac Sim and Omniverse largescale reinforcement learning and advanced VisionLanguageAction AI models The resulting robots powered by NVIDIA Thor target manufacturing logistics and retail sectors
quantumzeitgeist.com
June 12, 2025 at 4:05 PM
Humanoid Robotics and NVIDIA Partner to Accelerate Humanoid Robot Development
Read more:
https://quantumzeitgeist.com/humanoid-robotics-and-nvidia-partner-to-accelerate-humanoid-robot-development/
Read more:
https://quantumzeitgeist.com/humanoid-robotics-and-nvidia-partner-to-accelerate-humanoid-robot-development/