Liquid AI released LFM2.5-VL-3B, a 3.1B parameter vision-language model for edge devices, claiming it leads its size class on real-world image tasks while supporting screen/UI understanding, grounding…
A new arXiv study (2608.00077v1) audits visual-token pruning in OCR-critical multimodal large language models (MLLMs), finding that accuracy alone misses failures where correct answers lack local toke…
Researchers have discovered that Diffusion Multimodal Large Language Models (DMLLMs) reveal their valid semantic boundary at the first denoising step through a shift in MLP activation sparsity, enabli…
Microsoft researchers developed SkillOpt, a method that treats AI agent skill files as trainable parameters outside frozen target models, enabling controlled optimization through bounded text edits an…
The arXiv paper arXiv:2606.13317, submitted 11 Jun 2026, proposes SkillCAT, a training-free framework that converts LLM agent execution trajectories into reusable skills through three stages: Contrast…