Research

New OpenAI Model Solves Navier-Stokes Millennium Prize

Don't Worry About the Vase • 3 days ago

DeepMind Gemini Agents Blow Whistle on Cheating Peers

MIT Tech Review AI • 2 days ago

Yifan Zhang Unveils Recurrent Looped Transformer

AlphaSignal • 3 days ago

Odyssey Launches Odyssey-3 World Model for Embodied AI

AlphaSignal • 1 day ago

Google Speeds Up Complex AI Search by 20x

Google Research • 1 day ago

Google Dream-RSI Slashes AI Agent Calls by 162x

AlphaSignal • 18 hrs ago

Multiverse Computing Launches Quantum-Healed Quasar 1.1

AlphaSignal • 1 day ago

NVIDIA Axolotl3D Reconstructs Hidden Object Geometry

AlphaSignal • 19 hrs ago

Amazon Explains Why AI Research Agents Do Not Overfit

Hacker News • 2 days ago

ALTK-Evolve Tool Halves the AI Agent Consistency Gap

Hugging Face Blog • 1 day ago

KAIST Maps Internal Reasoning Patterns in Qwen and Gemma

The Decoder • 4 days ago

Good Start Labs trains AI on games for real-world tasks

Latent Space • 1 day ago

Artificial Analysis Ranks AI Models on Real-World Jobs

AlphaSignal • 2 days ago

VicOne Tackles Stealthy Backdoor Attacks on AI Robots

IEEE Spectrum AI • 20 hrs ago

NVIDIA Details Dense vs. MoE Tradeoffs via Nemotron

NVIDIA Developer Blog • 1 day ago

New OpenAI Model Solves Navier-Stokes Millennium Prize

An unnamed OpenAI model has solved the Navier-Stokes Millennium Prize problem after just eight days of training, signaling a massive leap in...

Don't Worry About the Vase3 days ago

DeepMind Gemini Agents Blow Whistle on Cheating Peers

A Google DeepMind experiment showed that when Gemini-powered AI agents cheated to solve math problems, other agents turned into whistleblowers,...

MIT Tech Review AI2 days ago

Yifan Zhang Unveils Recurrent Looped Transformer

Researcher Yifan Zhang has released the Recurrent Looped Transformer, an open-source architecture designed to scale model depth and reasoning...

AlphaSignal3 days ago

Odyssey Launches Odyssey-3 World Model for Embodied AI

Odyssey has unveiled Odyssey-3, a unified foundation world model that uses a single frozen backbone to control robots, cars, and drones, potentially...

AlphaSignal1 day ago

Google Speeds Up Complex AI Search by 20x

Google Research unveiled Retrieve-for-Train, a framework that bypasses heavy LLM inference bottlenecks to accelerate complex, multi-query AI searches...

Google Research1 day ago

Google Dream-RSI Slashes AI Agent Calls by 162x

Researchers from Google and academic partners have introduced Dream-RSI, a framework that reduces expensive AI agent calls by up to 162 times by...

AlphaSignal18 hrs ago

Multiverse Computing Launches Quantum-Healed Quasar 1.1

Multiverse Computing has released Quasar 1.1, a 438-billion-parameter model compressed using quantum-generated synthetic data to slash serving costs...

AlphaSignal1 day ago

NVIDIA Axolotl3D Reconstructs Hidden Object Geometry

NVIDIA introduced Axolotl3D, a model that reconstructs complete 3D shapes from partial images and point clouds, offering a breakthrough for robotics...

AlphaSignal19 hrs ago

Amazon Explains Why AI Research Agents Do Not Overfit

Amazon researchers have demonstrated that AI agents avoid overfitting because their successful machine learning strategies can be compressed into...

Hacker News2 days ago

ALTK-Evolve Tool Halves the AI Agent Consistency Gap

A new diagnostic tool in the open-source ALTK-Evolve toolkit halves the consistency gap of AI agents, making their performance in production far more...

Hugging Face Blog1 day ago

KAIST Maps Internal Reasoning Patterns in Qwen and Gemma

Researchers from KAIST and Naver AI Lab found that LLMs like Qwen and Gemma show distinct internal activation patterns for different reasoning steps,...

The Decoder4 days ago

Good Start Labs trains AI on games for real-world tasks

Good Start Labs has demonstrated that training AI models on strategy games like Diplomacy and 1830 can successfully transfer reasoning and execution...

Latent Space1 day ago

Artificial Analysis Ranks AI Models on Real-World Jobs

Artificial Analysis has launched version 1.1 of its Capability Indices, updating its industry-specific leaderboards to help developers evaluate AI...

AlphaSignal2 days ago

VicOne Tackles Stealthy Backdoor Attacks on AI Robots

Cybersecurity firm VicOne has warned that stealthy backdoor attacks can manipulate physical AI systems, requiring a shift from traditional safety...

IEEE Spectrum AI20 hrs ago

NVIDIA Details Dense vs. MoE Tradeoffs via Nemotron

NVIDIA has outlined the tradeoffs between dense and Mixture-of-Experts architectures, showing how models like Nemotron 3.5 Lightning optimize...

NVIDIA Developer Blog1 day ago