每天自动聚合 AI 领域最新动态
Language-model agents are usually trained by reinforcement learning from one reward per episode, and privileged self-...
Long-form thinking traces can substantially improve the multi-step reasoning performance of large language models (LL...
Multimodal large language models (MLLMs) struggle with spatial, embodied, physical, and temporal reasoning. We hypoth...
Modern AI models are aligned through post-training to adapt them to downstream tasks. Recent work shows that fine-tun...
Attention Residuals (AttnRes) enables content-dependent retrieval over network depth, which can be viewed as single-s...
Vision-and-Language Navigation (VLN) has largely focused on a single agent following a single instruction, yet many r...
When solving complex spatial reasoning tasks, ordinary humans naturally revisit the scenes from task-relevant viewpoi...
Large Language Models (LLMs) increasingly ship with explicit "thinking modes", yet their counterpart, "no-thinking", ...
Personal Graphical User Interface (GUI) agents rely on interaction history to infer what a user wants from ambiguous ...
As text-to-SQL systems move beyond demonstrations toward real-world deployment, ambiguity in user questions becomes a...
We reconstruct high-fidelity 3D scene meshes from posed multi-view images without per-scene optimization, across scal...
Terminal agents are moving beyond local operations toward completing real-world professional tasks. Short trajectorie...