See2Think: Do Multimodal Models Really Use Intermediate Visual States?

  • 类型：arxiv
  • 标识：2607.26769
  • 链接：https://arxiv.org/abs/2607.26769
  • 主分类：multimodal
  • 形态：benchmark
  • 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：Evaluating representative proprietary and open-source multimodal models, it is found that visual reasoning is strongly model- and environment-dependent, with no single setting consistently dominating across tasks.
  • OpenAlex ID：W7171872055
  • OpenAlex DOI：10.48550/arxiv.2607.26769
  • DOI：10.48550/arxiv.2607.26769
  • DOI来源：OpenAlex
  • 开放获取：green
  • 开放获取链接：https://doi.org/10.48550/arxiv.2607.26769
  • OpenAlex更新：2026-08-26
  • 副分类：evaluation
  • 待LLM分类：否
  • 标题中文：See2Think：多模态模型真的使用了中间视觉状态吗？
  • TLDR中文：对代表性闭源与开源多模态模型的评测表明，视觉推理强依赖于模型与环境，没有任何单一设置能在所有任务上持续占优。
  • 来源文件：
  • /inbox/tom/_candidates/2026-07-31-agent-rag-longcontext-candidates.json
  • /inbox/tom/_candidates/2026-08-01-agent-rag-longcontext-candidates.json
  • /inbox/tom/_candidates/2026-08-02-agent-rag-longcontext-candidates.json
  • /inbox/tom/_candidates/2026-08-03-agent-rag-longcontext-candidates.json
  • /inbox/tom/_candidates/2026-08-03-agent-memory-tool-use-candidates.json
  • [S2 enrich]
  • [OpenAlex backfill]