[论文] CRANE: Post-Hoc Counterfactual Retrieval for Multimodal Classifiers

论文概要

研究领域: ML 作者: Franco Rugolon, Ioanna Miliou, Panagiotis Papapetrou 发布时间: 2026-10-22 arXiv: 2610.31594

中文摘要

尚无针对多模态输入分类器的后验反事实解释方法。现有方法独立应用模态特异性归因，无法产生连贯解释。CRANE是首个后验的、模型无关的多模态反事实生成框架，从目标类检索真实训练示例作为候选解释，在构造上确保合理性。在四个异构数据集和三种融合策略上，多模态共识检索（MC-R）通过交集各模态邻域集成为最强策略，在四个数据集中的三个上取得最佳或并列最佳结果，同时将所有解释锚定在联合训练分布中。

原文摘要

No post-hoc method has been designed for classifiers that learn from multimodal inputs. Existing explainability methods apply modality-specific attribution independently to each input, failing to yield coherent explanations. We present CRANE, the first post-hoc, model-agnostic framework for nearest-neighbour multimodal counterfactual generation that retrieves real training examples from the target class as candidate explanations, ensuring plausibility by construction. Across four heterogeneous datasets and three fusion strategies, Multimodal Consensus Retrieval (MC-R), which intersects per-modality neighbour sets, emerges as the strongest multimodal strategy, achieving the best or joint-best outcome on three of four datasets while grounding every explanation in the joint training distribut...

#论文 #arXiv #ML #小凯