今日从 arXiv 订阅中筛选 10 篇论文。

Arxiv ID2608.26902幻觉翻译2608.26902

⚡ Tether the Subject, Release the Scene: Query-Aware Memory Routing for Long-Horizon Autoregressive Video Generation

Tether the Subject, Release the Scene: Query-Aware Memory Routing for Long-Horizon Autoregressive Video Generation
Arxiv ID2608.27065幻觉翻译2608.27065

⚡ Video-OPSD: Exploiting Privileged Visual Evidence for On-Policy Self-Distillation in Video LLMs

Video-OPSD: Exploiting Privileged Visual Evidence for On-Policy Self-Distillation in Video LLMs
Arxiv ID2608.26684幻觉翻译2608.26684

⚡ Reason in the Words You Speak: Idiolectal Paraphrasing Off-Policy Traces for Reasoning Distillation in VideoLLMs

Reason in the Words You Speak: Idiolectal Paraphrasing Off-Policy Traces for Reasoning Distillation in VideoLLMs
Arxiv ID2608.27073幻觉翻译2608.27073

⚡ SpatialCrafter: Single Image World Modeling with Generative 3D Proxies

SpatialCrafter: Single Image World Modeling with Generative 3D Proxies
Arxiv ID2608.26671幻觉翻译2608.26671

⚡ RECAP-Forcing: Retaining Content Appearances for Long Video Generation

RECAP-Forcing: Retaining Content Appearances for Long Video Generation
Arxiv ID2608.26794幻觉翻译2608.26794

⚡ Ring Forcing: Towards Precise Long-Term Memory for Autoregressive Video Diffusion

Ring Forcing: Towards Precise Long-Term Memory for Autoregressive Video Diffusion
Arxiv ID2608.27395幻觉翻译2608.27395

⚡ LeVJEPA: Efficient & Scalable Video Pretraining without the Heuristics

LeVJEPA: Efficient & Scalable Video Pretraining without the Heuristics
Arxiv ID2608.26809幻觉翻译2608.26809

⚡ Thinking on Shots: Consistent Multi-Shot Video Editing with Agentic Reasoning

Thinking on Shots: Consistent Multi-Shot Video Editing with Agentic Reasoning
Arxiv ID2608.27417幻觉翻译2608.27417

⚡ Retrieval Heads Meet Vision: Uncovering How VLMs Locate and Extract Visual Information

Retrieval Heads Meet Vision: Uncovering How VLMs Locate and Extract Visual Information
Arxiv ID2608.27226幻觉翻译2608.27226

⚡ DINOcular: Self-Supervised Visuospatial Representations

DINOcular: Self-Supervised Visuospatial Representations

自动生成于 2026-08-29 · 基于 arXiv Daily Digest