RelateAnything: Real-Time Open-Vocabulary Relation Prediction From Any Inputs

  • 类型：arxiv
  • 标识：2609.12552
  • 链接：https://arxiv.org/abs/2609.12552
  • 主分类：llm-infra
  • 形态：method
  • 被引：0
  • 被引来源：Semantic Scholar + OpenAlex
  • S2被引：0
  • OpenAlex被引：0
  • 影响力被引：0
  • TLDR：This work presents RelateAnything, a 53M-parameter model taking an image and regions from any source and returning scored relations over a predicate vocabulary supplied at inference as strings, a 53M-parameter model taking an image and regions from any source and returning scored relations over a predicate vocabulary supplied at inference as strings.
  • OpenAlex ID：W7212832108
  • OpenAlex DOI：10.48550/arxiv.2609.12552
  • DOI：10.48550/arxiv.2609.12552
  • DOI来源：OpenAlex
  • 开放获取：green
  • 开放获取链接：https://arxiv.org/pdf/2609.12552
  • OpenAlex更新：2026-09-18
  • 待LLM分类：否
  • 标题中文：RelateAnything: 任意输入的实时开放词汇关系预测
  • TLDR中文：本文提出 RelateAnything，一个 53M 参数的模型，输入一张图像和来自任意来源的区域，针对推理时以字符串形式提供的谓词词汇返回带分数的关系。
  • 来源文件：
  • /inbox/tom/_candidates/2026-09-17-agent-rag-longcontext-candidates.json
  • [OpenAlex backfill]
  • [S2 enrich]