提出新检索框架，解决多模态多跳问答中的重复信息问题。
Subtraction Gets You More: Gap-Aware Retrieval for Multimodal Multi-Hop QA
- 通过嵌入层隐式重写查询，动态选择检索策略。
- 在MultimodalQA上实现40.3%的宏平均性能提升。
- 适合需要跨模态推理与抗噪声检索的任务场景。
在多模态多跳问答中，我们关注初始检索阶段的两个任务：（1）证据集补全，即根据上下文检索缺失证据；（2）序列池构建，从零开始迭代生成前K个候选池。现有迭代检索框架常受语义锚定影响，先前获取的证据会困住检索器，导致实体中心的冗余。为此，我们提出GRAIL（基于间隙感知的自适应隐式定位检索），在嵌入层面直接进行隐式查询重写。通过上下文减法式查询引导，GRAIL在组合式跨模态推理上表现优异；而加法嵌入更新则在局部信息聚合上更具优势。根据任务类型动态路由查询，该混合框架在MultimodalQA上实现了40.3%的宏平均性能提升。大量实验表明，序列化GRAIL以更优且抗噪的方式检索，通过迭代间隙感知优化显著扩展了搜索范围。
原文摘要 · Abstract (English)
In multimodal multi-hop question answering, we focus on the initial retrieval stage via two distinct tasks: (1) evidence set completion, retrieving missing evidence given context, and (2) sequential pool construction, iteratively building the top-$K$ pool from the scratch. Under these settings, we point out that conventional iterative retrieval frameworks often suffer from Semantic Anchoring, where previously fetched evidence traps the retriever and yields entity-centric redundancy. To break this trap, we propose GRAIL (Gap-aware Retrieval via Adaptive Implicit Localization), a paradigm that performs implicit query rewriting directly at the embedding level. By context-subtractive query steering, GRAIL excels at compositional cross-modal reasoning, while additive embedding updates show strength on localized information aggregation. By dynamically routing queries based on task type, our Hybrid Framework achieves a 40.3% macro-averaged performance gain on MultimodalQA. Extensive evaluations demonstrate that sequential GRAIL retrieves in a superior, noise-resilient manner, significantly expanding the search horizon through iterative gap-aware optimization.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。