UniVR: Thinking in Visual Space for Unified Visual Reasoning Paper • 2607.12800 • Published 6 days ago • 28
KeyFrame-Compass: Towards Comprehensive Evaluation of Keyframe-Conditioned Video Generation Paper • 2607.14202 • Published 5 days ago • 36
LongStraw: Long-Context RL Beyond 2M Tokens under a Fixed GPU Budget Paper • 2607.14952 • Published 4 days ago • 174
view article Article Native-speed vLLM transformers modeling backend hmellor, lysandre • 12 days ago • 56
view article Article Distillation in 2026 (so far): which frontier models use it and how sergiopaniego • 11 days ago • 17
Video-Oasis: Rethinking Evaluation of Video Understanding Paper • 2603.29616 • Published 18 days ago • 64
Ideas Have Genomes: Benchmarking Scientific Lineage Reasoning and Lineage-Grounded Idea Generation Paper • 2607.08758 • Published 11 days ago • 39
Jet-Long: Efficient Long-Context Extension with Dynamic Bifocal RoPE Paper • 2607.07740 • Published 12 days ago • 23
Face-Depth 3D — Datasets Collection Collection of High-Quality Human Face 3D Depth Reconstruction Datasets • 4 items • Updated 9 days ago • 1
OpenDetection & Captioning — Datasets Collection An object detection and dense captioning dataset built using automated pipelines with RF-DETR, Qwen3.5, and other vision models. • 9 items • Updated 8 days ago • 2
DeepSeek v4 Flash — (Qwen35)(Distilled) Collection Collection of DeepSeek v4 Flash reasoning-distilled, (Qwen3.5) • 4 items • Updated 10 days ago • 1
PerceptionRubrics: Calibrating Multimodal Evaluation to Human Perception Paper • 2606.28322 • Published 24 days ago • 42
view article Article ScarfBench: Benchmarking AI Agents for Enterprise Java Framework Migration ibm-research • 19 days ago • 24
view article Article Kog Laneformer 2B: The Latency-First Model Behind Kog Inference Engine kogai • 25 days ago • 32
v2.0 — GLM-5.1/DeepSeek (Distilled) Collection Collection of GLM-5.1 or DeepSeek reasoning-distilled, Qwen3.5 • 4 items • Updated 10 days ago • 1