Latest stories
FM-Bench: New Benchmark Shows Managerial Behavior, Not Scale, Drives Long-Horizon LLM Agents
A new benchmark from Hugging Face, FM-Bench, tests LLM agents managing a football club over 20 simulated years, revealing that man...
New Diagnostics and Action-Conditioned Objectives Improve Latent World Model Planning
Researchers propose diagnostics to measure whether latent distances reflect true task progress in JEPA-style world models, and int...
Hugging Face Researchers Introduce Top-K Prompting to Boost Diverse Retrosynthesis Predictions
A new study from Hugging Face presents Top-K prompting and plausibility-aware training for single-step retrosynthesis, achieving s...
Stampli Uses OpenAI's Codex and ChatGPT Work to Cut Launch Time from Weeks to Days
Stampli, an accounts payable automation company, leveraged OpenAI's Codex and ChatGPT Work to compress weeks of launch production...
New Data-Free Method Traces Language Model Lineage via Weight Signatures
Researchers introduce a passive, data-free technique that verifies whether open-weight language model checkpoints share ancestry b...
SPADE: Self-Play Framework Lets LLMs Design Their Own Training Environments
Hugging Face researchers introduce SPADE, a self-play reinforcement learning framework where a language model acts as both environ...
SemaPLC: Verification-Gated Agent Harness Boosts PLC Code Generation Reliability
SemaPLC, a new agent harness from Midea AI, validates PLC code through external compilation and live runtime execution, achieving...
Hugging Face Introduces SemComp-Bench to Test Whether Video Generators Actually Complete Tasks
A new benchmark from Hugging Face, SemComp-Bench, evaluates video generation models on semantic task completion, measuring both ou...
Co-RL: Multi-Agent RL Enables Unsupervised Reasoning Without Labels
Hugging Face researchers introduce Co-RL, a multi-agent reinforcement learning framework where models learn from peer-derived rewa...
Hugging Face Unveils Zetta: A Closed-Loop Harness for Self-Evolving Robots
Zetta, a new closed-loop embodied harness from Hugging Face, enables robots to evolve runtime critics and recovery skills online,...
OpenAI Reaffirms Zero Data Retention for API Customers, Previews Private Safety Processing
OpenAI reiterates its Zero Data Retention commitment for eligible API customers and introduces a preview of Private Safety Process...
OpenAI Reaffirms Zero Data Retention and Previews Private Safety Processing
OpenAI reiterates its Zero Data Retention commitment for eligible API customers and introduces Private Safety Processing, a new ap...