Loading video...

Video Failed to Load

Go Home

Excited to share ESI-BENCH, a benchmark for Embodied Spatial Intelligence! Most spatial reasoning benchmarks assume an oracle observer: the agent is given the right image, view, or 3D scene. But in the real world, the observer is also an actor. To understand space, agents must decide where to look,...

yining_hong's profile picture

Yining Hong

4,583 subscribers

51,990 views • 2 months ago •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Chatbots aren’t the revolution. They’re the distraction. Fei-Fei Li: “Language is a half-million-year-old luxury. Perception is a half-billion-year-old necessity.” Evolution didn’t optimize for conversation. It optimized for survival in three-dimensional space. Seeing threats, navigating obstacles, predicting what happens when you move. We’ve spent years celebrating AI that can write and summarize. But text processing is narrow. Spatial intelligence is fundamental. An agent that only reads prompts can’t function in a warehouse or a hospital. It needs to parse depth, understand physics, and act on what it sees in real time. We built AI that understands language. Now we’re building AI that understands space. Language models got the attention. Spatial intelligence gets the work done. The world runs on physics, not paragraphs. AI is learning to operate in it.
1:01

Chatbots aren’t the revolution. They’re the distraction. Fei-Fei Li: “Language is a half-million-year-old luxury. Perception is a half-billion-year-old necessity.” Evolution didn’t optimize for conversation. It optimized for survival in three-dimensional space. Seeing threats, navigating obstacles, predicting what happens when you move. We’ve spent years celebrating AI that can write and summarize. But text processing is narrow. Spatial intelligence is fundamental. An agent that only reads prompts can’t function in a warehouse or a hospital. It needs to parse depth, understand physics, and act on what it sees in real time. We built AI that understands language. Now we’re building AI that understands space. Language models got the attention. Spatial intelligence gets the work done. The world runs on physics, not paragraphs. AI is learning to operate in it.

Dustin

51,772 views • 6 months ago

Dr. Fei-Fei Li just called out the biggest blind spot in the entire AI industry. We have been building half of human intelligence. And calling it the finish line. Li: “If you look at human intelligence, it pretty much boils down to two buckets.” The first bucket is language. Symbolic reasoning. Communication. The ability to think in words and abstractions. That’s what every major AI lab has spent the last decade building. The second bucket is the one the industry has almost entirely ignored. Li: “We call that in AI spatial intelligence.” How humans and animals perceive, navigate, and interact with the three-dimensional physical world. How we reach for objects. How we move through space. How we build and manipulate physical reality. From painting masterpieces to constructing the pyramids, non-verbal spatial intelligence is what actually shapes the world. Language describes reality. Spatial intelligence acts on it. And the gap between those two things is the gap between a chatbot and a robot. Li: “When this technology is ready, the robotic revolution is gonna start. We’re already seeing that trend.” Every robot is a moving agent. Every moving agent requires spatial intelligence to function in the real world. The humanoid robots being deployed in factories right now are hitting the ceiling of what language models alone can power. Spatial intelligence is the unlock. But Li didn’t stop at robotics. Li: “From a geopolitics point of view, this is part of the technology that goes straight into weapons.” Autonomous drone swarms. Battlefield navigation. Physical target acquisition without human oversight. Every military application of AI that operates in the real world runs on spatial intelligence. The nation that masters the transition from static text to dynamic three-dimensional perception doesn’t just win the software race. It commands the physical battlefield. The AI arms race just broke out of the data center. It’s operating in three dimensions now.
3:08

Dr. Fei-Fei Li just called out the biggest blind spot in the entire AI industry. We have been building half of human intelligence. And calling it the finish line. Li: “If you look at human intelligence, it pretty much boils down to two buckets.” The first bucket is language. Symbolic reasoning. Communication. The ability to think in words and abstractions. That’s what every major AI lab has spent the last decade building. The second bucket is the one the industry has almost entirely ignored. Li: “We call that in AI spatial intelligence.” How humans and animals perceive, navigate, and interact with the three-dimensional physical world. How we reach for objects. How we move through space. How we build and manipulate physical reality. From painting masterpieces to constructing the pyramids, non-verbal spatial intelligence is what actually shapes the world. Language describes reality. Spatial intelligence acts on it. And the gap between those two things is the gap between a chatbot and a robot. Li: “When this technology is ready, the robotic revolution is gonna start. We’re already seeing that trend.” Every robot is a moving agent. Every moving agent requires spatial intelligence to function in the real world. The humanoid robots being deployed in factories right now are hitting the ceiling of what language models alone can power. Spatial intelligence is the unlock. But Li didn’t stop at robotics. Li: “From a geopolitics point of view, this is part of the technology that goes straight into weapons.” Autonomous drone swarms. Battlefield navigation. Physical target acquisition without human oversight. Every military application of AI that operates in the real world runs on spatial intelligence. The nation that masters the transition from static text to dynamic three-dimensional perception doesn’t just win the software race. It commands the physical battlefield. The AI arms race just broke out of the data center. It’s operating in three dimensions now.

Dustin

122,680 views • 5 months ago

World Labs CEO Dr. Fei-Fei Li explains says "world model" has become an overloaded term & explains what each kind of world model does: "Right now there are three ways of calling world models when it comes to spatial intelligence." "One is what I call a renderer, when the model puts beautiful pixels on the screen." "Another kind of world model is what we call a planner. That is more for machines, more for robots." "The third kind, which I think is the linchpin of the three, is a simulator." "A simulator could become a renderer. The simulator could become a planner. But this layer is a huge critical path to unlock spatial intelligence. And that's what World Labs is working on." Fei-Fei Li at Bloomberg Tech live with Emily Chang
2:08

World Labs CEO Dr. Fei-Fei Li explains says "world model" has become an overloaded term & explains what each kind of world model does: "Right now there are three ways of calling world models when it comes to spatial intelligence." "One is what I call a renderer, when the model puts beautiful pixels on the screen." "Another kind of world model is what we call a planner. That is more for machines, more for robots." "The third kind, which I think is the linchpin of the three, is a simulator." "A simulator could become a renderer. The simulator could become a planner. But this layer is a huge critical path to unlock spatial intelligence. And that's what World Labs is working on." Fei-Fei Li at Bloomberg Tech live with Emily Chang

a16z

112,090 views • 2 months ago

🔥In Magma, we talked a lot about spatial/temporal intelligence beyond verbal intelligencen as advocated by Dr. Fei-Fei Li. So how to interpret it? Today I am happy to announce a new demo Magma-Gaming: 👉 Rather than asking LLMs to write game code, we further ask the model to PLAY the game. A simple game like moving to the target in a 2D grid still requires precise action grounding and planning capability, yet challenges the most advanced VLMs and even GPT-4o-mini model. Magma, born with stronger spatial understanding and reasoning ability, significantly outperforms the counterparts and achieves much higher scores in zero-shot manner. This result pinpoints the huge potential of building multimodal agentic models endowed with both verbal and spatial inteligence! Also, I believe this simple demo gives you a better hints to understand what is spatial intelligence, and why it is important!
1:05

🔥In Magma, we talked a lot about spatial/temporal intelligence beyond verbal intelligencen as advocated by Dr. Fei-Fei Li. So how to interpret it? Today I am happy to announce a new demo Magma-Gaming: 👉 Rather than asking LLMs to write game code, we further ask the model to PLAY the game. A simple game like moving to the target in a 2D grid still requires precise action grounding and planning capability, yet challenges the most advanced VLMs and even GPT-4o-mini model. Magma, born with stronger spatial understanding and reasoning ability, significantly outperforms the counterparts and achieves much higher scores in zero-shot manner. This result pinpoints the huge potential of building multimodal agentic models endowed with both verbal and spatial inteligence! Also, I believe this simple demo gives you a better hints to understand what is spatial intelligence, and why it is important!

Jianwei Yang

17,940 views • 1 year ago

Excited to share our latest work on 🎧spatial audio-driven human motion generation. We aim to tackle a largely underexplored yet important problem of enabling virtual humans to move naturally in response to spatial audio—capturing not just what is heard, but also where the sound is coming from. To this end, we introduce the Spatial Audio-Driven Human Motion (SAM) dataset—the first comprehensive dataset featuring paired high-quality human motion and spatial audio recordings. For benchmarking, we develop a generative framework for human MOtion generation driven by SPAtial audio, termed MOSPA, which learns to synthesize realistic and diverse human motions conditioned on spatial audio input. We hope this research could provide a foundation for future research in spatial perception, virtual characters, and embodied AI. The dataset and model will be open-sourced soon. A big thank you to our intern, Shuyang Xu, for the wonderful collaboration! Congratulations, Shuyang! Project page: Paper: Video: #Animation #CG #CV #AIGC #DL #Deeplearning #Motion #Graphics #AI #GenerativeAI
3:58

Excited to share our latest work on 🎧spatial audio-driven human motion generation. We aim to tackle a largely underexplored yet important problem of enabling virtual humans to move naturally in response to spatial audio—capturing not just what is heard, but also where the sound is coming from. To this end, we introduce the Spatial Audio-Driven Human Motion (SAM) dataset—the first comprehensive dataset featuring paired high-quality human motion and spatial audio recordings. For benchmarking, we develop a generative framework for human MOtion generation driven by SPAtial audio, termed MOSPA, which learns to synthesize realistic and diverse human motions conditioned on spatial audio input. We hope this research could provide a foundation for future research in spatial perception, virtual characters, and embodied AI. The dataset and model will be open-sourced soon. A big thank you to our intern, Shuyang Xu, for the wonderful collaboration! Congratulations, Shuyang! Project page: Paper: Video: #Animation #CG #CV #AIGC #DL #Deeplearning #Motion #Graphics #AI #GenerativeAI

Zhiyang (Frank) Dou

14,610 views • 1 year ago

In their latest video, Boston Dynamics’s AI team explains how they make the Atlas humanoid perceive and interact with the world. Atlas uses an agile perception system to understand both the shape and context of objects in complex environments. Atlas combines 2D and 3D awareness, keypoint-based localization, and an object pose tracking system that fuses vision, kinematics, and object knowledge to handle occlusion and uncertainty. Accurate calibration ensures precise hand–eye coordination for reliable manipulation. The team is now working toward a unified model that merges perception and control – pushing beyond spatial AI toward physical intelligence.
6:34

In their latest video, Boston Dynamics’s AI team explains how they make the Atlas humanoid perceive and interact with the world. Atlas uses an agile perception system to understand both the shape and context of objects in complex environments. Atlas combines 2D and 3D awareness, keypoint-based localization, and an object pose tracking system that fuses vision, kinematics, and object knowledge to handle occlusion and uncertainty. Accurate calibration ensures precise hand–eye coordination for reliable manipulation. The team is now working toward a unified model that merges perception and control – pushing beyond spatial AI toward physical intelligence.

The Humanoid Hub

50,083 views • 1 year ago

What would a World Model look like if we start from a real embodied agent acting in the real world? It has to have: 1) A real, physically grounded and complex action space—not just abstract control signals. 2) Diverse, real-life scenarios and activities. Or in short: It has to be annoyingly complex—in both the action and vision space—to even get close to real life. We did an initial attempt: Whole-Body Conditioned Egocentric Video Prediction. In collaboration with @dans_t123 , Amir Bar, Yann LeCun , trevordarrell and Jitendra MALIK. (For more details, check: What we did is very simple: Predict Egocentric Video from human Actions (PEVA) - Given the past video and a future action represented by relative 3D body pose, PEVA predicts how the world looks next—from the first-person view. By conditioning on kinematic pose trajectories, structured by the joint hierarchy of the body, it learns how physical actions shape perception.
3:12

What would a World Model look like if we start from a real embodied agent acting in the real world? It has to have: 1) A real, physically grounded and complex action space—not just abstract control signals. 2) Diverse, real-life scenarios and activities. Or in short: It has to be annoyingly complex—in both the action and vision space—to even get close to real life. We did an initial attempt: Whole-Body Conditioned Egocentric Video Prediction. In collaboration with @dans_t123 , Amir Bar, Yann LeCun , trevordarrell and Jitendra MALIK. (For more details, check: What we did is very simple: Predict Egocentric Video from human Actions (PEVA) - Given the past video and a future action represented by relative 3D body pose, PEVA predicts how the world looks next—from the first-person view. By conditioning on kinematic pose trajectories, structured by the joint hierarchy of the body, it learns how physical actions shape perception.

Yutong Bai

179,017 views • 1 year ago

Fei-Fei Li ( Fei-Fei Li ) beautifully explains Robotics. She defines robotics not by form, like humanoids or cars, but by function: they are any "embodied machines" that must perceive, understand, and act within a physical, 3D space. This core requirement is "spatial intelligence," the unifying principle of all robotics, allowing them to perform tasks and even collaborate with humans. Throughout all of human history, we have been confined to a single, shared reality: the "physical Earth 3D world." This singularity has been our only playground. However, new technologies that combine 3D generation and reconstruction are shattering this limitation. We can now create "infinite universes"—a multiverse of digital worlds for countless purposes, from training robots to enabling creativity, travel, and storytelling. This leap from one physical world to an infinite multiverse unlocks boundless possibilities for human imagination and interaction. Video from a16z
1:25

Fei-Fei Li ( Fei-Fei Li ) beautifully explains Robotics. She defines robotics not by form, like humanoids or cars, but by function: they are any "embodied machines" that must perceive, understand, and act within a physical, 3D space. This core requirement is "spatial intelligence," the unifying principle of all robotics, allowing them to perform tasks and even collaborate with humans. Throughout all of human history, we have been confined to a single, shared reality: the "physical Earth 3D world." This singularity has been our only playground. However, new technologies that combine 3D generation and reconstruction are shattering this limitation. We can now create "infinite universes"—a multiverse of digital worlds for countless purposes, from training robots to enabling creativity, travel, and storytelling. This leap from one physical world to an infinite multiverse unlocks boundless possibilities for human imagination and interaction. Video from a16z

Rohan Paul

41,441 views • 6 months ago

Fei-Fei Li ( Fei-Fei Li ) beautifully explains Robotics. She defines robotics not by form, like humanoids or cars, but by function: they are any "embodied machines" that must perceive, understand, and act within a physical, 3D space. This core requirement is "spatial intelligence," the unifying principle of all robotics, allowing them to perform tasks and even collaborate with humans. Throughout all of human history, we have been confined to a single, shared reality: the "physical Earth 3D world." This singularity has been our only playground. However, new technologies that combine 3D generation and reconstruction are shattering this limitation. We can now create "infinite universes"—a multiverse of digital worlds for countless purposes, from training robots to enabling creativity, travel, and storytelling. This leap from one physical world to an infinite multiverse unlocks boundless possibilities for human imagination and interaction. Video from a16z
1:25

Fei-Fei Li ( Fei-Fei Li ) beautifully explains Robotics. She defines robotics not by form, like humanoids or cars, but by function: they are any "embodied machines" that must perceive, understand, and act within a physical, 3D space. This core requirement is "spatial intelligence," the unifying principle of all robotics, allowing them to perform tasks and even collaborate with humans. Throughout all of human history, we have been confined to a single, shared reality: the "physical Earth 3D world." This singularity has been our only playground. However, new technologies that combine 3D generation and reconstruction are shattering this limitation. We can now create "infinite universes"—a multiverse of digital worlds for countless purposes, from training robots to enabling creativity, travel, and storytelling. This leap from one physical world to an infinite multiverse unlocks boundless possibilities for human imagination and interaction. Video from a16z

Rohan Paul

51,174 views • 3 months ago

🚨🚨Can agents earn money, run a business, or even self-organize a society in the physical social world? 🤖🤖 Can agents learn continually to survive and thrive in embodied environments, like how human babies grow? 👶 Super excited to introduce SimWorld, an open-ended simulator of LLM agents in infinite, realistic embodied worlds. SimWorld features 3 key designs: 1⃣Open-ended realistic world simulation - built on Unreal Engine 5, with accurate physical social dynamics - 100+ built-in environments (city, island, wilderness ...) - language-controllable procedural generation - text-to-3D asset generation 2⃣Native interface for LLM/VLM agents - Gym-like agent-environment interaction APIs - plug in any LLMs/VLMs (GPTs, Gemini, Qwen ...) - rich multi-modal perception - open-vocabulary natural-language action outputs 3⃣Diverse physical and social reasoning scenarios - long-horizon embodied reasoning - multi-agent collaboration / competition - easily customizable for any reasoning tasks SimWorld is fully open-sourced, with a hope to become a foundational infrastructure for real-world agent research across disciplines: robotics, economy, public health, education, etc. Project website + more details in the thread👇 ...1/
1:10

🚨🚨Can agents earn money, run a business, or even self-organize a society in the physical social world? 🤖🤖 Can agents learn continually to survive and thrive in embodied environments, like how human babies grow? 👶 Super excited to introduce SimWorld, an open-ended simulator of LLM agents in infinite, realistic embodied worlds. SimWorld features 3 key designs: 1⃣Open-ended realistic world simulation - built on Unreal Engine 5, with accurate physical social dynamics - 100+ built-in environments (city, island, wilderness ...) - language-controllable procedural generation - text-to-3D asset generation 2⃣Native interface for LLM/VLM agents - Gym-like agent-environment interaction APIs - plug in any LLMs/VLMs (GPTs, Gemini, Qwen ...) - rich multi-modal perception - open-vocabulary natural-language action outputs 3⃣Diverse physical and social reasoning scenarios - long-horizon embodied reasoning - multi-agent collaboration / competition - easily customizable for any reasoning tasks SimWorld is fully open-sourced, with a hope to become a foundational infrastructure for real-world agent research across disciplines: robotics, economy, public health, education, etc. Project website + more details in the thread👇 ...1/

Lianhui Qin

65,128 views • 8 months ago

Terminal-Bench 2.0 went from ~25% → 80% in four months and became the standard eval for frontier CLI agents. Now, TB3 is in the works. I talked to Alex Shaw about what happens when model capabilities climb faster than we can measure them. His answer: the benchmark factory (Harbor Framework)— infrastructure to develop hard, representative evals at the pace that the frontier moves. As Alex put it: "we need a thousand times more benchmarks than we have right now." 00:23 - How quickly models hill-climbed TB2 01:46 - What rapid progress reveals about benchmarks vs. real-world capability 03:28 - What made Terminal-Bench stick 04:58 - Why the terminal is the right abstraction for agentic AI 07:14 - How TB2 maintains task quality at scale 09:23 - Managing benchmark integrity in a benchmaxxing world 10:47 - Harbor: from experiment to benchmark factory 12:19 - What Harbor does that nothing else did 14:37 - The invariants: what won't change as agent evals evolve 16:55 - The benchmark Alex most wants to see built 18:18 - The ideal human-in-the-loop task creation flywheel 20:32 - How to contribute to Terminal-Bench 3.0
22:17

Terminal-Bench 2.0 went from ~25% → 80% in four months and became the standard eval for frontier CLI agents. Now, TB3 is in the works. I talked to Alex Shaw about what happens when model capabilities climb faster than we can measure them. His answer: the benchmark factory (Harbor Framework)— infrastructure to develop hard, representative evals at the pace that the frontier moves. As Alex put it: "we need a thousand times more benchmarks than we have right now." 00:23 - How quickly models hill-climbed TB2 01:46 - What rapid progress reveals about benchmarks vs. real-world capability 03:28 - What made Terminal-Bench stick 04:58 - Why the terminal is the right abstraction for agentic AI 07:14 - How TB2 maintains task quality at scale 09:23 - Managing benchmark integrity in a benchmaxxing world 10:47 - Harbor: from experiment to benchmark factory 12:19 - What Harbor does that nothing else did 14:37 - The invariants: what won't change as agent evals evolve 16:55 - The benchmark Alex most wants to see built 18:18 - The ideal human-in-the-loop task creation flywheel 20:32 - How to contribute to Terminal-Bench 3.0

vincent sunn chen

11,563 views • 4 months ago

World Labs CEO Dr. Fei-Fei Li says AI must change how we teach and evaluate students: "AI must change learning. AI must change K-16 learning." "The most precious resource of our entire world is human capital." "When we have gotten the technology that can answer standardized tests... when AI can do better than an average human, it's not about humans are bad. It's about we need to change the education system." "We need to change how we evaluate. We need to change the way we empower teachers to educate the next generation of students where they can use these tools, be empowered, and do things that we can never imagine." "All of the kids today should not be scared of AI. They should feel the human agency to lead AI, to use AI in the right way, and to use AI to make the impact that they want to make for the world." Fei-Fei Li at Bloomberg Tech live with Emily Chang
1:35

World Labs CEO Dr. Fei-Fei Li says AI must change how we teach and evaluate students: "AI must change learning. AI must change K-16 learning." "The most precious resource of our entire world is human capital." "When we have gotten the technology that can answer standardized tests... when AI can do better than an average human, it's not about humans are bad. It's about we need to change the education system." "We need to change how we evaluate. We need to change the way we empower teachers to educate the next generation of students where they can use these tools, be empowered, and do things that we can never imagine." "All of the kids today should not be scared of AI. They should feel the human agency to lead AI, to use AI in the right way, and to use AI to make the impact that they want to make for the world." Fei-Fei Li at Bloomberg Tech live with Emily Chang

a16z

81,767 views • 2 months ago

Introducing WorldView 3D, a first-of-its-kind satellite tasking product that delivers up-to-date 3D ground truth anywhere on Earth. From GPS-denied autonomy and intelligence analysis to consumer-grade mapping and disaster response, the world's most important missions increasingly depend on accurate, up-to-date 3D terrain. But for years, keeping 3D maps current at global scale hasn't been possible. WorldView 3D changes that. Built on Vantor’s trusted spatial foundation, our new tasking capability empowers customers to task and receive updated 3D data within 24 hours of image collection. For missions that require greater fidelity, they can also produce high-definition 3D maps anywhere in the world. 3D is the future of spatial intelligence, and Vantor is leading the change. Our trusted spatial foundation already includes 100 million sq km of the world mapped inhighly accurate 3D. Now customers can update the locations that matter most with the speed, scale, and fidelity required for critical missions. Read the full announcement:
0:35

Introducing WorldView 3D, a first-of-its-kind satellite tasking product that delivers up-to-date 3D ground truth anywhere on Earth. From GPS-denied autonomy and intelligence analysis to consumer-grade mapping and disaster response, the world's most important missions increasingly depend on accurate, up-to-date 3D terrain. But for years, keeping 3D maps current at global scale hasn't been possible. WorldView 3D changes that. Built on Vantor’s trusted spatial foundation, our new tasking capability empowers customers to task and receive updated 3D data within 24 hours of image collection. For missions that require greater fidelity, they can also produce high-definition 3D maps anywhere in the world. 3D is the future of spatial intelligence, and Vantor is leading the change. Our trusted spatial foundation already includes 100 million sq km of the world mapped inhighly accurate 3D. Now customers can update the locations that matter most with the speed, scale, and fidelity required for critical missions. Read the full announcement:

Vantor

16,030 views • 1 month ago

Dr Fei-Fei-Li explains with a simple example how everyday household chores are so extremely difficult for Robots. "If you tell a robot to open the top drawer and watch out for the vase, this is actually a really hard task for robots." because the robot must ground language into the real world. Words like "top", "drawer", and "vase" are abstract. The system has to map them to 3D locations, objects, and relations in a noisy scene. This requires robust perception, object recognition, and spatial reasoning under uncertainty. The robot also lacks human commonsense. "Watch out" implies predicting consequences, estimating clearances, and understanding that vases are fragile. Encoding such priors, like how heavy a drawer is or how a vase might tip, is very complex and difficult without rich world knowledge. Learning the behavior from rewards is tough. The success signal is very sparse here, so naive exploration almost never stumbles on a full success sequence. This makes policy learning sample inefficient and brittle, especially when the environment changes between training and deployment. A sparse reward situation is when the agent only gets a success signal at the very end, and gets little or no feedback along the way. If a robot must open a drawer without hitting a vase, it might get reward only if the drawer ends up open and the vase is intact. Every partial try before that looks the same to the learner, reward equals 0. --- From "DSAI by Dr. Osbert Tay" YT channel
0:48

Dr Fei-Fei-Li explains with a simple example how everyday household chores are so extremely difficult for Robots. "If you tell a robot to open the top drawer and watch out for the vase, this is actually a really hard task for robots." because the robot must ground language into the real world. Words like "top", "drawer", and "vase" are abstract. The system has to map them to 3D locations, objects, and relations in a noisy scene. This requires robust perception, object recognition, and spatial reasoning under uncertainty. The robot also lacks human commonsense. "Watch out" implies predicting consequences, estimating clearances, and understanding that vases are fragile. Encoding such priors, like how heavy a drawer is or how a vase might tip, is very complex and difficult without rich world knowledge. Learning the behavior from rewards is tough. The success signal is very sparse here, so naive exploration almost never stumbles on a full success sequence. This makes policy learning sample inefficient and brittle, especially when the environment changes between training and deployment. A sparse reward situation is when the agent only gets a success signal at the very end, and gets little or no feedback along the way. If a robot must open a drawer without hitting a vase, it might get reward only if the drawer ends up open and the vase is intact. Every partial try before that looks the same to the learner, reward equals 0. --- From "DSAI by Dr. Osbert Tay" YT channel

Rohan Paul

342,627 views • 8 months ago

The teams shipping AI agents right now are bleeding money on the dumbest possible expense: teaching a 400B-parameter model to read a file name. Every time an AI agent needs to "see" something today, it routes an image through a frontier model. OCR, object detection, checking if a button exists on screen. You're paying GPT-4o or Claude pricing for tasks that require perception, not reasoning. One agent workflow processing a few thousand screenshots per day can burn through more on vision calls than on the actual thinking. Perceptron's Isaac is 2B parameters. Built by the team that created Meta's Chameleon multimodal models. On perceptive benchmarks, it matches or beats models 50x its size. The VQA, OCR, and object detection scores are competitive with models running on infrastructure that costs orders of magnitude more. The MCP wrapper is the distribution play. One install command and every Claude Code agent can offload vision tasks to a model that runs on a single consumer GPU. The agent keeps its reasoning in the frontier model and routes perception to a specialist. That split is how you get vision-heavy agent workflows from "technically possible but expensive" to "cheap enough to run on everything." This is the same pattern that won in every other compute-intensive stack. General-purpose handles orchestration. Specialists handle the heavy lifting. Graphics went through it. Audio went through it. Video encoding went through it. Vision in AI agents is next. The teams building agents that see 10,000 images a day will care about this before anyone else does.
0:38

The teams shipping AI agents right now are bleeding money on the dumbest possible expense: teaching a 400B-parameter model to read a file name. Every time an AI agent needs to "see" something today, it routes an image through a frontier model. OCR, object detection, checking if a button exists on screen. You're paying GPT-4o or Claude pricing for tasks that require perception, not reasoning. One agent workflow processing a few thousand screenshots per day can burn through more on vision calls than on the actual thinking. Perceptron's Isaac is 2B parameters. Built by the team that created Meta's Chameleon multimodal models. On perceptive benchmarks, it matches or beats models 50x its size. The VQA, OCR, and object detection scores are competitive with models running on infrastructure that costs orders of magnitude more. The MCP wrapper is the distribution play. One install command and every Claude Code agent can offload vision tasks to a model that runs on a single consumer GPU. The agent keeps its reasoning in the frontier model and routes perception to a specialist. That split is how you get vision-heavy agent workflows from "technically possible but expensive" to "cheap enough to run on everything." This is the same pattern that won in every other compute-intensive stack. General-purpose handles orchestration. Specialists handle the heavy lifting. Graphics went through it. Audio went through it. Video encoding went through it. Vision in AI agents is next. The teams building agents that see 10,000 images a day will care about this before anyone else does.

Aakash Gupta

55,978 views • 4 months ago

Right now the main paradigm that we think of agents in is chatting back and forth, but the biggest use of tokens will come from agents that are just always on running in the background doing work for us, or ones triggered from a workflow. Agents will be working 24/7 in our workflows processing data, reviewing and generating documents, moving data between systems, writing code, accelerating decision making steps, and more. In Claude's new Managed Agents feature, in a couple minutes you can wire up an agent that can read contracts when they come into Box to review them, and then assign a task in Linear with the critical information from the contract. But this could have been any workflow, like reviewing documents for client onboarding, invoice processing, M&A due-diligence, data extraction pipelines, and millions of other use-cases. And integrating data across any system. This is only possible when you can have long-running agents that can complete real work in the background, accurately. Agents have the ability to execute code safely, leverage tools, access a compute sandbox, and connect across systems is clearly the architecture of the future. The industry is now making it easier and easier for enterprises to build and deploy these agents.
0:43

Right now the main paradigm that we think of agents in is chatting back and forth, but the biggest use of tokens will come from agents that are just always on running in the background doing work for us, or ones triggered from a workflow. Agents will be working 24/7 in our workflows processing data, reviewing and generating documents, moving data between systems, writing code, accelerating decision making steps, and more. In Claude's new Managed Agents feature, in a couple minutes you can wire up an agent that can read contracts when they come into Box to review them, and then assign a task in Linear with the critical information from the contract. But this could have been any workflow, like reviewing documents for client onboarding, invoice processing, M&A due-diligence, data extraction pipelines, and millions of other use-cases. And integrating data across any system. This is only possible when you can have long-running agents that can complete real work in the background, accurately. Agents have the ability to execute code safely, leverage tools, access a compute sandbox, and connect across systems is clearly the architecture of the future. The industry is now making it easier and easier for enterprises to build and deploy these agents.

Aaron Levie

17,543 views • 4 months ago

Introducing ALE-Bench, ALE-Agent! Towards Automating Long-Horizon Algorithm Engineering for Hard Optimization Problems Blog: Paper: ALE-Bench is a coding benchmark primarily focused on hard optimization (NP-hard) problems. We developed this benchmark with AtCoder Inc., a leading coding contest platform company. What makes ALE-Bench unique is its focus on hard optimization problems that demand long-horizon and creative reasoning. It’s open-ended, in the sense that true optima are out of reach (NP-hard) and scores can continuously improve. We believe this benchmark has the potential to become one of the key benchmarks for reasoning and coding in the next generation. ALE-Agent is our end-to-end agent that we specifically designed for this challenging domain. In fact, our ALE-Agent has already built an impressive track record in the wild! In May 2025, our agent participated in a live AtCoder Heuristic Competition (AHC), alongside 1,000 other participants in real-time. AHC is considered to be one of the most challenging coding competitions in this domain. Our ALE-Agent achieved an impressive ranking of 21st out of 1,000 human participants in the competition (top 2%), marking a turning point for AI discovery of solutions to hard optimization problems with a wide spectrum of important real world applications such as logistics, routing, packing, factory production planning, power-grid balancing. We look forward to applying this technology to real industrial optimization opportunities. Building on the insights from this study, Sakana AI will continue to tackle the challenge of developing AI with even greater algorithm engineering capabilities. ALE-Bench Dataset: ALE-Bench Code: This research was conducted in collaboration with AtCoder Inc. (AtCoder). We are deeply grateful for their outstanding expertise and contributions in optimization and algorithms, which were invaluable in providing data, analyzing results, and enabling our AI agent’s participation in their contests.
0:54

Introducing ALE-Bench, ALE-Agent! Towards Automating Long-Horizon Algorithm Engineering for Hard Optimization Problems Blog: Paper: ALE-Bench is a coding benchmark primarily focused on hard optimization (NP-hard) problems. We developed this benchmark with AtCoder Inc., a leading coding contest platform company. What makes ALE-Bench unique is its focus on hard optimization problems that demand long-horizon and creative reasoning. It’s open-ended, in the sense that true optima are out of reach (NP-hard) and scores can continuously improve. We believe this benchmark has the potential to become one of the key benchmarks for reasoning and coding in the next generation. ALE-Agent is our end-to-end agent that we specifically designed for this challenging domain. In fact, our ALE-Agent has already built an impressive track record in the wild! In May 2025, our agent participated in a live AtCoder Heuristic Competition (AHC), alongside 1,000 other participants in real-time. AHC is considered to be one of the most challenging coding competitions in this domain. Our ALE-Agent achieved an impressive ranking of 21st out of 1,000 human participants in the competition (top 2%), marking a turning point for AI discovery of solutions to hard optimization problems with a wide spectrum of important real world applications such as logistics, routing, packing, factory production planning, power-grid balancing. We look forward to applying this technology to real industrial optimization opportunities. Building on the insights from this study, Sakana AI will continue to tackle the challenge of developing AI with even greater algorithm engineering capabilities. ALE-Bench Dataset: ALE-Bench Code: This research was conducted in collaboration with AtCoder Inc. (AtCoder). We are deeply grateful for their outstanding expertise and contributions in optimization and algorithms, which were invaluable in providing data, analyzing results, and enabling our AI agent’s participation in their contests.

Sakana AI

237,195 views • 1 year ago

SITUATION EXPLAINED: A new generation of AI labs are betting that world models are the path to general intelligence that can interact with the real world. We had General Intuition CEO Pim de Witte on MTS to explain what world models are, and why intuitively world models offer a clearer path towards general intelligence than LLMs. For humans, text is a technology for compressing information about the unfolding, three-dimensional world around us and communicating it to other people. When LLMs reason about the world, they do so by reconstructing the world through text. World models work by modeling the three-dimensional world at the architecture level and then allowing agents to interact with that world. Unlike an LLM, which understands space and time through text, world models perform spatial and temporal reasoning natively. General Intuition hopes to use world models to create embodied AI—machines that can reason and act in the real world. But other companies are also pushing the frontier in this space. Odyssey ML today announced a multi-agent world model, Agora-1. Their launch video shows four participants interacting with a real-time world model simulation of GoldenEye, a first-person shooter for the N64. A playable preview of the GoldenEye simulation is available on their site.
1:31

SITUATION EXPLAINED: A new generation of AI labs are betting that world models are the path to general intelligence that can interact with the real world. We had General Intuition CEO Pim de Witte on MTS to explain what world models are, and why intuitively world models offer a clearer path towards general intelligence than LLMs. For humans, text is a technology for compressing information about the unfolding, three-dimensional world around us and communicating it to other people. When LLMs reason about the world, they do so by reconstructing the world through text. World models work by modeling the three-dimensional world at the architecture level and then allowing agents to interact with that world. Unlike an LLM, which understands space and time through text, world models perform spatial and temporal reasoning natively. General Intuition hopes to use world models to create embodied AI—machines that can reason and act in the real world. But other companies are also pushing the frontier in this space. Odyssey ML today announced a multi-agent world model, Agora-1. Their launch video shows four participants interacting with a real-time world model simulation of GoldenEye, a first-person shooter for the N64. A playable preview of the GoldenEye simulation is available on their site.

MTS

11,641 views • 2 months ago

The video does not show an “unarmed Good Samaritan.” It shows a chaotic physical struggle where a man later found to be armed intervenes, is pepper-sprayed, goes to the ground, and is then actively wrestled as ICE agents move in to regain control. As the suspect stands up during the struggle, agents reasonably perceive a firearm coming into play—and the first shot is fired. Under settled self-defense law, officers are not required to wait until a gun is fully raised or fired. Imminence includes the moment a weapon is accessed during resistance. Freeze-frame activists keep asking “can you see the gun?” That’s the wrong legal question. The standard is reasonable perception in real time, including shouted warnings, movement, and the dynamics of a fight. This was a rapidly evolving use-of-force encounter, not an execution. The law does not require officers to gamble with their lives.
2:00

The video does not show an “unarmed Good Samaritan.” It shows a chaotic physical struggle where a man later found to be armed intervenes, is pepper-sprayed, goes to the ground, and is then actively wrestled as ICE agents move in to regain control. As the suspect stands up during the struggle, agents reasonably perceive a firearm coming into play—and the first shot is fired. Under settled self-defense law, officers are not required to wait until a gun is fully raised or fired. Imminence includes the moment a weapon is accessed during resistance. Freeze-frame activists keep asking “can you see the gun?” That’s the wrong legal question. The standard is reasonable perception in real time, including shouted warnings, movement, and the dynamics of a fight. This was a rapidly evolving use-of-force encounter, not an execution. The law does not require officers to gamble with their lives.

Andrew Branca Show

351,308 views • 6 months ago