Multimodal Continuous Reasoning via Asymmetric Mutual Variational Learning

  • 类型:arxiv
  • 标识:2607.00461
  • 链接:https://arxiv.org/abs/2607.00461
  • 主分类:multimodal
  • 形态:method
  • 被引:1
  • 被引来源:Semantic Scholar
  • S2被引:1
  • OpenAlex被引:0
  • 影响力被引:0
  • TLDR:AMVL is instantiate in a latent-integrated MLLM and it consistently outperforms strong discrete and latent-reasoning baselines, improving the average score on the complex BLINK benchmark by +10.83 and achieving gains of up to +32.00 on individual reasoning tasks, with analyses confirming improved latent-space stability.
  • OpenAlex ID:W7167126791
  • OpenAlex DOI:10.48550/arxiv.2607.00461
  • DOI:10.48550/arxiv.2607.00461
  • DOI来源:OpenAlex
  • 开放获取:green
  • 开放获取链接:https://doi.org/10.48550/arxiv.2607.00461
  • OpenAlex更新:2026-07-19
  • 副分类:llm-infra
  • 待LLM分类:否
  • 标题中文:基于非对称互变分学习的多模态连续推理
  • TLDR中文:AMVL 在潜空间集成的 MLLM 中实例化,持续优于多种强离散和潜空间推理基线,在复杂 BLINK 基准上平均得分提升 +10.83,单个推理任务最高提升 +32.00,分析也确认了潜空间稳定性的改善。
  • 来源文件
  • /inbox/tom/_candidates/2026-07-02-agent-rag-longcontext-candidates.json
  • [S2 enrich]
  • [OpenAlex backfill]