A team of engineers at the Micro Air Vehicle Lab at Delft University of Technology won a high-profile Abu Dhabi first-person-view (FPV) drone race in November 2025 with an AI-piloted quadcopter, the f…
VIDRAFT released Darwin-180B-RSI, a 180-billion-parameter reasoning model trained with a weight-level Recursive Self-Improvement framework that updates model weights directly without human-annotated c…
Researchers at the University of Zurich released LittleLearner, a family of language models (0.6B, 1.3B, and 5B parameters) trained from scratch on LittleCurriculum, an 88B-token corpus filtered to U.…
A new study from researchers at the University of Zurich finds that reinforcement learning with verifiable rewards surpasses supervised fine-tuning for legal neural machine translation, though enhance…
Reddit is using its own AI tools, including large language models, to combat AI-generated content and spam on its platform, blocking 23 million spam views daily and reducing user spam exposure by 20 p…