Papers Videos Whiteboards Open Problems Pricing Log in Sign up

Papers Whiteboards Videos Open Problems Pricing Log in Sign up

Generate videos that make complex ideas click

Sign Up to Generate Custom Videos

Religion and AI as Distributed Meaning Systems

B-spline Policy: Accelerating Manipulation with Continuous Action Representations

Requential Coding: Compressing Models by Teaching Themselves

Trident: Adaptive Scheduling for Heterogeneous Multimodal Data Pipelines

Measuring Intelligence Beyond Human Scale

Soofi S: A Sovereign Foundation Model for German and English

Lumo-2: Teaching Robots to Predict the Physics of Action

From Chaos to Structure: The Unreasonable Power of Random Matrices

Billions of Sketches Reveal Hidden Cultural Variation in Human Concepts

ProxyPose: Tracking 3D Motion Without 3D Models

Single-Rollout Asynchronous Optimization for Agentic Reinforcement Learning

Video Generation Models as Universal Vision Foundation Models

Vision Pretraining for Dense Spatial Perception

Wan-Streamer v0.2: Higher Resolution, Same Latency

KAT-Coder-V2.5: Infrastructure-Driven Code Agents

RoboDojo: When Generalist Robot Policies Meet Reality

Collective Cognition in Hybrid Groups: A Network Science Synthesis

Gemma 4: Rethinking Multimodal AI from the Ground Up

The Final Strand: Proving Burau is Faithful for Four Braids

When Fireballs Don't Cooperate: A Statistical Null Result Story

LLM-as-a-Verifier: Scaling Verification for Smarter AI Agents

Kani: A Model Checker for Rust

NeuroCogMap Reveals Cognitive Organization of Large Language Models

A Mathematical Introduction to Diffusion Models

AdaJEPA: Adaptive Latent World Models That Learn While They Plan

When LLMs Develop Languages: Symbolic Communication for Efficient Multi-Agent Reasoning

LeVLJEPA: Vision-Language Learning Without Negatives

One Layer Is Enough: Rethinking RL Training in Transformers

GPC: Large-Scale Generative Pretraining for Transferable Motor Control

Hierarchical Global Attention: Scaling Transformers to 64K Tokens

The Verification Horizon: No Silver Bullet for Coding Agent Rewards

Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent

The Shift to Agentic AI: Evidence from Codex

Ask, Don't Judge: Binary Questions for Interpretable LLM Evaluation

What is it Like to Be a Bot: The S3Q Theory of Consciousness

SPIRAL: Teaching Language Models to Search, Explore, and Synthesize

You Don't Need to Run Every Eval

Autodata: Teaching AI Agents to Generate Their Own Training Data

Data Centers Actually Lowered Your Electric Bill

OpenThoughts-Agent: Data Recipes for Agentic Models

Qwen-AgentWorld: Language World Models for General Agents

Unlimited OCR Works: Breaking the Long Document Bottleneck

Teaching Language Models to Think Like Real Humans in Persuasion Games

SC3-Eval: Self-Consistent Video Generation for Robot Policy Evaluation

Freeing the Law with LOCUS: A Local Ordinance Corpus for the United States

BabelTele: When LLMs Speak in Code Humans Can't Read

Do as I Do: Learning Dexterous Manipulation from Everyday Human Videos

Five Generations of TPU: Architectural Stability at Supercomputer Scale

VibeThinker-3B: Exploring the Frontier of Verifiable Reasoning in Small Language Models

Fearless Concurrency on the GPU

Chrome Extension