每天自动聚合 AI 领域最新动态
Training and evaluating robot policies in the real world is costly and difficult to scale. We introduce SimFoundry, a...
Artificial intelligence is driving a revolution in scientific discovery, accelerating everything from hypothesis gene...
We introduce an axiomatic evaluation framework for latent thought representations in LLMs, comprising metrics that ar...
We present Qwen-Image-2.0-RL, a post-training pipeline that applies reinforcement learning from human feedback (RLHF)...
Vision-language models (VLMs) are increasingly deployed in consumer, medical, financial, and enterprise applications....
Reinforcement learning (RL) post-training improves the reward alignment of flow-based generators, but often degrades ...
Web-agent benchmarks overwhelmingly measure depth -- pinning one obscure answer behind a chain of constraints -- whil...
Knowledge-based Visual Question Answering (KB-VQA) requires models to combine image understanding with external knowl...
We study whether we can learn novel manipulation skills from human actions to a bi-manual robot with parallel gripper...
Large language models (LLMs) can make scientific software easier to use. However, a general model does not automatica...
I describe my solution to the LeHome Challenge 2026, an ICRA 2026 competition on bimanual garment folding. The system...
Vision-Language-Action (VLA) models can generalize across diverse manipulation tasks, but their imitation-learning-ba...