Remote KV caching for long-context LLM serving can severely degrade latency unless the store is sharded by parallel rank, ...

Streaming vision-language models (VLMs) must continuously integrate information from the past while remaining responsive to ...

OneWM unifies causal training of world and action modeling, showing that learning action-conditioned dynamics improves ...

Velox accelerates long video rollouts by routing attention via geometric frustum slabs, reducing latency while preserving ...

W e p l a c e d f o u r G e m i n i 4 A r g o n a g e n t s i n a s a n d b o x e d M i n e c r a f t s u r v i v a l w o r l ...

Muse Spark Open model. A 816B parameter model with 42B Active parameters. BF16 and NVFP4 Weights.

The method converts pre-trained transformers into elastic models that adapt continuously to varying computational budgets by ...

Hi everyone, first author here! Personalizing diffusion models on synthetic images can produce oversaturated colors and excessive texture. We trace this to inflated classifier-free guidance, ...

Multi-teacher on-policy distillation (MOPD) is used in two settings. In common-domain composition, several teachers score ...

Author here. WebFovea is a vision-based web agent that placed 2nd in the WebRetriever Challenge 2026 (57.0/100). It works on live websites through their own UI: screenshots in, clicks and keystrokes ...

This is an automated message from the Librarian Bot. I found the following papers similar to this paper.

Index-Translate-35B-A3B-preview is the 35B-total / 3B-active mixture-of-experts multilingual text model of the Index-Translate family, built on Qwen3.5. It supports text translation across 150 ...