OpenVLHarness: Real World Visual Agentic Tool-Use with Multimodal Memory Paper • 2610.26860 • Published 26 days ago • 103
GameHorizon Suite: Multi-Horizon Data and Evaluation in Gameplay Paper • 2609.25001 • Published Sep 21 • 46
Vidu S2: Real-Time Interactive, Editable, and Spatial Video Generation Paper • 2609.11638 • Published Sep 10 • 223
StateM: Reaching 95.3% Raw Accuracy, or a \$15 Frontier Run, on Terminal-Bench 2.1 via Harness Scaling Paper • 2608.15089 • Published Aug 15 • 106
FinMME: Benchmark Dataset for Financial Multi-Modal Reasoning Evaluation Paper • 2505.24714 • Published May 30, 2025 • 17