AI Agent Papers2026
Lets a pretrained multimodal LLM interpret navigation requests and orchestrate motion via tool calls for generalist robot navigation across unfamiliar scenes. Key features: an agent harness with Navigation Skills, a unified visual-point interface, task-progress tracking, and tool-based motion execution without navigation-specific fine-tuning.
Uses large-scale, mixed-task reinforcement learning to drive self-improvement of multimodal foundation models. Key features include fully-asynchronous large-batch RL (1,568 prompts × 16 rollouts, up to 3.7B tokens/update, 1M context), groupwise agentic grading, and sparse MoE architectures.
AI Image2026
A drop-in GGUF text encoder (Qwen3‑VL‑8B‑Instruct) with its refusal direction projected out for use with stable-diffusion.cpp; drastically reduces chat refusals while leaving the 7B image denoiser and visual outputs essentially unchanged.
AI Video Papers2026
Converts pretrained multimodal LLMs into a System One decision interface that outputs structured, software-friendly decisions for video (temporal spans or multiple-choice answers); uses a shared scoring head and cached/joint inference to raise temporal-grounding mIoU and drastically speed up repeated queries.