Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Ritual makes inference a native precompile, so reasoning and the action it triggers settle as one operation. Humans don't have gaps between a thought and action, and neither should an agent.

joshsimenhoff's profile picture

Josh (❖,❖)

6,644 subscribers

10,744 Aufrufe • vor 1 Monat •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Humans are weird. Two skis, and we act as if we can fly. 🎿 Action 6 caught it all. ⚠️ Professional operation. Do not imitate. 🎥：ficekooo #DJI #OsmoAction6
0:19

Humans are weird. Two skis, and we act as if we can fly. 🎿 Action 6 caught it all. ⚠️ Professional operation. Do not imitate. 🎥：ficekooo #DJI #OsmoAction6

DJI

16,876 Aufrufe • vor 1 Monat

MolmoAct2 is landing in LeRobot! Ai2's open Action Reasoning Model combines a Molmo2-ER vision-language backbone with a flow-matching continuous action expert to predict robot action chunks from images, language instructions, and proprioceptive state. An open robot foundation model built for real-world control, with strong out-of-the-box performance and easy fine-tuning in LeRobot. Pick-and-place inference running on NVIDIA DGX Spark! Blog: Paper: Thanks to Ai2 Jiafei Duan Haoquan Fang
0:11

MolmoAct2 is landing in LeRobot! Ai2's open Action Reasoning Model combines a Molmo2-ER vision-language backbone with a flow-matching continuous action expert to predict robot action chunks from images, language instructions, and proprioceptive state. An open robot foundation model built for real-world control, with strong out-of-the-box performance and easy fine-tuning in LeRobot. Pick-and-place inference running on NVIDIA DGX Spark! Blog: Paper: Thanks to Ai2 Jiafei Duan Haoquan Fang

LeRobot

24,727 Aufrufe • vor 1 Monat

How can we responsibly deploy agentic AI robots in real-world environments? As we close the gap between AI reasoning and physical action, developing a new safety framework is key to advancing helpful robots. 🤖
0:13

How can we responsibly deploy agentic AI robots in real-world environments? As we close the gap between AI reasoning and physical action, developing a new safety framework is key to advancing helpful robots. 🤖

Google DeepMind

47,887 Aufrufe • vor 9 Monaten

Polyurethane foam in action—fills gaps, lifts concrete, and hardens to a durable, concrete-like finish..
0:11

Polyurethane foam in action—fills gaps, lifts concrete, and hardens to a durable, concrete-like finish..

Harri

12,448 Aufrufe • vor 1 Monat

Introducing bunnyOS: the small-model agent and inference platform. Small language models can now reason, plan, use tools, and take action. We are building the infrastructure to unlock them. Litepaper in bio.
0:13

Introducing bunnyOS: the small-model agent and inference platform. Small language models can now reason, plan, use tools, and take action. We are building the infrastructure to unlock them. Litepaper in bio.

BunnyOS

314,634 Aufrufe • vor 3 Tagen

In Abu Dhabi, the parties never stop and neither does the action.
0:19

In Abu Dhabi, the parties never stop and neither does the action.

Fast & Furious

34,095 Aufrufe • vor 1 Jahr

got cloud inference working! here i have smolvla running on a remote 4090 on prime intellect with a thin client running on my local pc sending observations to the remote server and receiving action chunks back. round trip latency is varying between 320-470ms! the 4090 instance I'm using is $0.62/hr and not even one of the cheaper ones, a raspberry pi v5 is < $100 and about $25 for a v3, which works fine as a client
0:13

got cloud inference working! here i have smolvla running on a remote 4090 on prime intellect with a thin client running on my local pc sending observations to the remote server and receiving action chunks back. round trip latency is varying between 320-470ms! the 4090 instance I'm using is $0.62/hr and not even one of the cheaper ones, a raspberry pi v5 is < $100 and about $25 for a v3, which works fine as a client

Jack Vial

12,992 Aufrufe • vor 5 Monaten

"An unexpected encounter at the zoo: a tender moment between humans and pandas"...
0:15

"An unexpected encounter at the zoo: a tender moment between humans and pandas"...

Intresting nature 🌿🪐

22,560 Aufrufe • vor 1 Monat

All For One Quirk Skill Intros! Special Action "Plunder" Deals the final blow on a downed opponent and steals their Quirk! Switch between stolen Quirks using Special Action! #MHUR
0:17

All For One Quirk Skill Intros! Special Action "Plunder" Deals the final blow on a downed opponent and steals their Quirk! Switch between stolen Quirks using Special Action! #MHUR

MY HERO ULTRA RUMBLE

75,230 Aufrufe • vor 2 Jahren

Introduce CoT-VLA – Visual Chain-of-Thought reasoning for Robot Foundation Models! 🤖 By leveraging next-frame prediction as visual chain-of-thought reasoning, CoT-VLA uses future prediction to guide action generation and unlock large-scale video data for training. #CVPR2025
0:11

Introduce CoT-VLA – Visual Chain-of-Thought reasoning for Robot Foundation Models! 🤖 By leveraging next-frame prediction as visual chain-of-thought reasoning, CoT-VLA uses future prediction to guide action generation and unlock large-scale video data for training. #CVPR2025

Qingqing Zhao

48,207 Aufrufe • vor 1 Jahr

I thought fishing needed a bit more action: so I gave some fish boxing gloves! Sea Sniffers will have +100 fish and fish-likes, all catchable (and punchable) 😋
0:13

I thought fishing needed a bit more action: so I gave some fish boxing gloves! Sea Sniffers will have +100 fish and fish-likes, all catchable (and punchable) 😋

Richard Lems ☕

40,796 Aufrufe • vor 7 Monaten

Neurons don’t physically touch—they communicate across tiny gaps called synapses. Signals travel as electrical impulses within a neuron and chemical neurotransmitters between neurons, enabling thought, memory, and movement
0:16

Neurons don’t physically touch—they communicate across tiny gaps called synapses. Signals travel as electrical impulses within a neuron and chemical neurotransmitters between neurons, enabling thought, memory, and movement

Science girl

71,962 Aufrufe • vor 2 Monaten

Mike Brown has stuff ready for Mikal Bridges as a secondary option. Slot to slot flip and toss back triggers Sabonis elbow flash. First look here is the split cut action between Murray/Ellis. Warriors switch and take away the back door (option 1). Ellis flows right into an away screen for DeRozan.
0:13

Mike Brown has stuff ready for Mikal Bridges as a secondary option. Slot to slot flip and toss back triggers Sabonis elbow flash. First look here is the split cut action between Murray/Ellis. Warriors switch and take away the back door (option 1). Ellis flows right into an away screen for DeRozan.

DJ Zullo

22,687 Aufrufe • vor 11 Monaten

Rockets don't differ between Israeli and Gazan civilians. Neither does Hamas. Everyone is a target.
0:14

Rockets don't differ between Israeli and Gazan civilians. Neither does Hamas. Everyone is a target.

Israel Defense Forces

765,329 Aufrufe • vor 2 Jahren

Marriage, as God designed it, is the covenant union of one man and one woman. 💍 A union between a MAN and a WOMAN.
0:13

Marriage, as God designed it, is the covenant union of one man and one woman. 💍 A union between a MAN and a WOMAN.

GYATA BA🦁 🇬🇭🇬🇧

11,676 Aufrufe • vor 11 Tagen

SECRETARY RUBIO: We have complied with the law and what it requires. This is an action by the President to eliminate a real threat to the safety and security of the United States and to our allies.
0:13

SECRETARY RUBIO: We have complied with the law and what it requires. This is an action by the President to eliminate a real threat to the safety and security of the United States and to our allies.

Department of State

27,303 Aufrufe • vor 4 Monaten

Current Vision-Language-Action (VLA) paradigms in autonomous driving primarily rely on Imitation Learning (IL), which introduces inherent challenges such as distribution shift and causal confusion. Online Reinforcement Learning offers a promising pathway to address these issues through trial-and-error learning. However, applying online reinforcement learning to VLA models in autonomous driving is hindered by inefficient exploration in continuous action spaces. MindDrive, a VLA framework comprising a large language model (LLM) with two distinct sets of LoRA parameters. The one LLM serves as a Decision Expert for scenario reasoning and driving decision-making, while the other acts as an Action Expert that dynamically maps linguistic decisions into feasible trajectories. Paper Title: MindDrive: A Vision-Language-Action Model for Autonomous Driving via Project: Link:
0:12

Current Vision-Language-Action (VLA) paradigms in autonomous driving primarily rely on Imitation Learning (IL), which introduces inherent challenges such as distribution shift and causal confusion. Online Reinforcement Learning offers a promising pathway to address these issues through trial-and-error learning. However, applying online reinforcement learning to VLA models in autonomous driving is hindered by inefficient exploration in continuous action spaces. MindDrive, a VLA framework comprising a large language model (LLM) with two distinct sets of LoRA parameters. The one LLM serves as a Decision Expert for scenario reasoning and driving decision-making, while the other acts as an Action Expert that dynamically maps linguistic decisions into feasible trajectories. Paper Title: MindDrive: A Vision-Language-Action Model for Autonomous Driving via Project: Link:

AI Bites | YouTube Channel

43,496 Aufrufe • vor 4 Monaten