今日从 arXiv 订阅中筛选 10 篇论文。

Arxiv ID2606.24051幻觉翻译2606.24051

⚡ DriveStack-VLA: BEV VLA driving

DriveStack-VLA: BEV VLA driving
Arxiv ID2606.24759幻觉翻译2606.24759

⚡ UniDrive: Risk understanding grounding

Arxiv ID2606.23938幻觉翻译2606.23938

⚡ Neuro-Symbolic Drive: Rule-grounded driving VLA

Neuro-Symbolic Drive: Rule-grounded driving VLA
Arxiv ID2606.24422幻觉翻译2606.24422

⚡ EgoSAT: Egocentric video reasoning benchmark

EgoSAT: Egocentric video reasoning benchmark
Arxiv ID2606.24152幻觉翻译2606.24152

⚡ Autonomous Video Generation: Counterfactual world models

Autonomous Video Generation: Counterfactual world models
Arxiv ID2606.24292幻觉翻译2606.24292

⚡ ActiveScope: MLLM perception correction

ActiveScope: MLLM perception correction
Arxiv ID2606.24597幻觉翻译2606.24597

⚡ Qwen-AgentWorld: Language world models

Qwen-AgentWorld: Language world models
Arxiv ID2606.24539幻觉翻译2606.24539

⚡ PointVG-R: Geometric reasoning grounding

PointVG-R: Geometric reasoning grounding
Arxiv ID2606.24231幻觉翻译2606.24231

⚡ FlowR2A: Reward-to-action driving planning

FlowR2A: Reward-to-action driving planning
Arxiv ID2606.24726幻觉翻译2606.24726

⚡ SER: Semantic evidence for video reasoning

自动生成于 2026-06-24 · 基于 arXiv Daily Digest