mmBERT: a modern multilingual encoder Collection mmBERT is trained on 3T tokens from over 1800 languages, showing SoTA scores on benchmarks and exceptional low-resource performance • 16 items • Updated Sep 9, 2025 • 54
ARIS: Autonomous Research via Adversarial Multi-Agent Collaboration Paper • 2605.03042 • Published 29 days ago • 124
OccuBench: Evaluating AI Agents on Real-World Professional Tasks via Language World Models Paper • 2604.10866 • Published Apr 13 • 66
Open Pangram Collection Open models and datasets based on Pangram's ICLR 2026 EditLens paper licensed for noncommercial use ONLY under CC BY-NC-SA 4.0 • 4 items • Updated Apr 24 • 15
PRBench: End-to-end Paper Reproduction in Physics Research Paper • 2603.27646 • Published Mar 29 • 29