3rd Place at CVPR 2026 CASTLE Challenge: Agentic Multi-View Long-Context Video Understanding via Hierarchical Knowledge Graph Retrieval Paper • 2606.01933 • Published Jun 1 • 1
Tencent WorkBuddy Bench: A Multi-Domain Coding-Agent Benchmark with Contamination-Resistant Task Construction Paper • 2607.20911 • Published 5 days ago • 24
JOSIE-2 4B Preview Collection The new J.O.S.I.E. version 2 family of personality-driven language models trained entirely on Apple Silicon • 8 items • Updated about 15 hours ago • 4
Environment-free Synthetic Data Generation for API-Calling Agents Paper • 2607.16900 • Published 10 days ago • 20
DataFlow-Harness: A Grounded Code-Agent Platform for Constructing Editable LLM Data Pipelines Paper • 2607.16617 • Published 10 days ago • 136
view article Article Welcome Inkling by Thinking Machines +2 burtenshaw, merve, pcuenq, ariG23498 • 13 days ago • 129
S1-Omni: A Unified Multimodal Reasoning Model for Scientific Understanding, Prediction, and Generation Paper • 2607.15686 • Published 11 days ago • 16