One Success Isn't Reliability: Thinkingbox, a Sandbox and Benchmark for Agents in Stateful Business Workflows Paper • 2608.19741 • Published 16 days ago • 12
PixRestore: Unified Image Restoration via Pixel Diffusion Transformer Paper • 2608.16793 • Published 19 days ago • 3
MoE-ViE: Mixture of Experts Vision Encoder for Efficient Image and Video Understanding Paper • 2608.17402 • Published 18 days ago • 18
A Frozen Pixel-Space Diffusion Model Can Guide Itself with Its Own Samples Paper • 2607.29122 • Published Jul 31 • 5
WorldDiT: A Unified Diffusion Architecture for World and Action Modeling Paper • 2607.23909 • Published Jul 27 • 7
MonkeyOCRv2: A Visual-Text Foundation Model for Document AI Paper • 2607.11562 • Published Jul 13 • 80
Xiaomi-Robotics-U0: Unified Embodied Synthesis with World Foundation Model Paper • 2607.11643 • Published Jul 13 • 45
Single-Rollout Asynchronous Optimization for Agentic Reinforcement Learning Paper • 2607.07508 • Published Jul 8 • 32
Sparse Delta Memory: Scaling the State of Linear RNNs through Sparsity Paper • 2607.07386 • Published Jul 8 • 13
HunyuanOCR-1.5: Making Lightweight OCR VLMs Faster and Better Paper • 2607.04884 • Published Jul 6 • 9
Unified Audio Intelligence Without Regressing on Text Intelligence Paper • 2607.05196 • Published Jul 6 • 24
view post Post 429 @alvarobartt published a step-by-step guide to deploy zai-org/GLM-5.2 on AMD GPUs, using newly released features in Microsoft Foundry. The FP8 model fits on a single node of MI300X GPUs, which cuts the bill in half vs. H100.https://alvarobartt.com/goal-glm-5.2-on-foundry/ See translation 👀 1 1 + Reply
Duration Aware Scheduling for ASR Serving Under Workload Drift Paper • 2603.11273 • Published Mar 11 • 3
Ultralytics YOLO26: Unified Real-Time End-to-End Vision Models Paper • 2606.03748 • Published Jun 2 • 24
Gemini Embedding 2: A Native Multimodal Embedding Model from Gemini Paper • 2605.27295 • Published May 26 • 25
Geometric Context Transformer for Streaming 3D Reconstruction Paper • 2604.14141 • Published Apr 15 • 38
A Frame is Worth One Token: Efficient Generative World Modeling with Delta Tokens Paper • 2604.04913 • Published Apr 6 • 12