Top AI Papers This Week
The AI paper with the strongest traction this week is PDRVLA: A Vision-Language-Action Model with Physical Dynamics Representations (arXiv:2610.37381), with 152 community upvotes on Hugging Face.
Each Monday we pull the week's AI papers from Hugging Face Daily Papers — the community's upvoted picks of new arXiv submissions — and surface the ten with the strongest traction, so the week's research signal is one page, not ten thousand papers.
PDRVLA: A Vision-Language-Action Model with Physical Dynamics Representations
Vision--Language--Action (VLA) models generate robot actions from visual observations and language instructions, but their understanding of physical dynamics remains limited. Video world models can capture scene dynamics, yet generating fut…
OmniShowcase: Omnidirectional Subject-Consistent Video Generation from Multi-View Images
Generating videos of a specific subject from a few reference images remains challenging when the camera reveals surfaces that are not visible in the input. Early subject-to-video (S2V) methods condition on a single reference image and must …
Efficient Streaming Audio-Visual Target Speaker Extraction for Real-World Acoustic Scenes
Streaming audio-visual target speaker extraction (AVTSE) is essential for latency-sensitive applications. However, current causal systems face two coupled gaps. First, standard benchmarks fix the number of speakers at two and ignore the nea…
Hy-Nomas: Towards Free-form Interactive World Generation
Interactive world generation requires models to follow successive commands while preserving relevant history and action consequences. We present Hy-Nomas, a video world model with an 8B diffusion backbone that supports free-form text instru…
Work Agents: Systems, Capabilities, and Optimization
As foundation models evolve from information generation to digital work execution, agents are increasingly capable of completing complex tasks across tools, applications, and workflows. However, existing research lacks a unified perspective…
Look Only When Needed: Certificate-Guided Visual Routing for Efficient Diffusion VLMs
Diffusion vision-language models (DVLMs) process their entire input at every denoising step. Since visual tokens make up 89-97% of this input on the benchmarks we study, most of the decoding computation is spent repeatedly processing the sa…
StreamVibe: Proactive Video Agents for Evolving User Intent
Vision-language models are shifting from offline passive understanding toward online proactive interaction. However, prevailing methods largely rely on explicit requests while miss the implicit user intent (e.g., "where should I click next?…
TAO: Tool-Aware Joint Scheduling for LLM-Based Agentic Serving
Agentic services alternate between model inference on clustered GPUs and external tools on distributed CPUs. In this scenario, existing systems (1) overlook tool-return readiness in service evaluation, wasting limited resources, and (2) use…
Rollback: Experience Backtracking in World Models for Policy Improvement
Action-conditioned world models offer a scalable alternative to costly real-world interaction for improving robot policies. However, existing world-model-based methods largely treat world models as forward-only surrogates of the physical en…
ChronoReplay: One Answer Interface Is Not Enough for Streaming Spatial Intelligence
Training a spatial VLM to answer which side an object lies on does not require it to preserve the metric displacement that determines that answer, while supervision for a distance estimate may leave categorical relations largely unconstrain…
Frequently Asked Questions
How are this week's top AI papers selected?
We collect the last seven days of Hugging Face Daily Papers — the community's upvoted picks of new arXiv submissions — and rank the ten with the most upvotes.
Are arXiv papers peer-reviewed?
No. arXiv is a preprint server — results have not necessarily been peer-reviewed. Treat these as the week's research conversation, not settled science.
Why rank by upvotes instead of citations?
Upvotes reflect interest within days of release. Citations take months to accumulate, so for a weekly chart they would all read zero.
Editor’s Notes
- Papers are drawn from Hugging Face Daily Papers, which curates noteworthy new arXiv submissions (cs.AI and cs.CL among them).
- Ranking is by community upvotes on Hugging Face over the trailing seven days.
- arXiv is a preprint server — papers are not peer-reviewed.
- Snapshot taken every Monday at 00:00 UTC.
Last updated: 2026-11-02