PulseAugur
中
实时 06:25:12
English(EN) SPARC: Separating Perception And Reasoning Circuits for Test-time Scaling of VLMs

SPARC框架解耦视觉语言模型（VLM）的感知和推理，以实现增强的扩展性

作者 PulseAugur 编辑部 · [1 个来源] ·

研究人员开发了SPARC，一个旨在提高视觉语言模型（VLM）性能和可扩展性的新框架。SPARC将视觉感知与推理分离，允许在推理过程中动态调整token预算。这种模块化方法能够独立优化感知和推理电路，从而提高效率和准确性，尤其是在分布外场景中。SPARC在具有挑战性的视觉推理任务上展示了显著的性能提升，优于单体基线模型，并降低了计算成本。 AI

影响 这种视觉语言模型（VLM）架构的模块化方法可能带来更高效、更适应复杂视觉推理任务的模型。

排序理由 该集群包含一篇详细介绍视觉语言模型（VLM）新框架的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

SPARC框架解耦视觉语言模型（VLM）的感知和推理，以实现增强的扩展性

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
该集群包含一篇详细介绍视觉语言模型（VLM）新框架的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
paper, model release
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
111 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

  1. arXiv cs.CL TIER_1 English(EN) · Niccolo Avogaro, Nayanika Debnath, Li Mi, Thomas Frick, Junling Wang, Zexue He, Hang Hua, Konrad Schindler, Mattia Rigotti ·

    SPARC：分离感知和推理电路以实现视觉语言模型（VLMs）的测试时扩展

    arXiv:2602.06566v3 Announce Type: replace-cross Abstract: Despite recent successes, test-time scaling -- i.e., dynamically expanding the token budget during inference as needed -- remains brittle for vision-language models (VLMs). Unstructured visual reasoning chains entangle per…

相关实体

相关话题