发表机构
The Hong Kong University of Science and Technology (Guangzhou); The Hong Kong University of Science and Technology; Nanyang Technological University; Singapore Institute of Manufacturing Technology, A*STAR; Lingnan University（香港科技大学（广州）; 香港科技大学; 南洋理工大学; 新加坡制造技术研究所，新加坡科技研究局; 岭南大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究多模态情感分析中缺失模态是否总需修复，提出SIEVE方法，通过比较直接预测与修复分支，从样本损失差距得经验充分性信号，经证据门路由输入，与修复无关，实验证明其能改进修复主干并接近最优值。
AI 中文摘要
现有的多模态情感分析（MSA）中缺失模态的方法通常遵循先修复的范式。我们重新审视这一假设并提出疑问：每个缺失模态都应该被修复吗？样本级的神谕分析表明并非总是如此：全模态输入仅对一小部分样本最优，某些样本偏好每个模态子集。这表明添加或修复模态不一定总能改善预测，且每个模态的效用取决于样本。基于此，我们提出SIEVE，将‘是否修复’转化为样本级可学习的决策。它比较直接预测分支和修复分支，从样本损失差距中得出经验充分性信号，通过证据门路由输入。SIEVE与修复无关，可在任何修复模块之上即插即用。在CMU - MOSI和IEMOCAP上的实验表明，SIEVE在不同缺失率下持续改进代表性修复主干，并接近样本级双分支可实现的最优值。
英文摘要
Existing methods for multimodal sentiment analysis (MSA) under missing modalities usually follow a repair-first paradigm. We revisit this assumption and ask: \emph{should every missing modality be repaired?} A per-sample oracle analysis shows the answer is not always: full-modality input is optimal for only a small fraction of samples, and every modality subset is preferred by some samples. These results suggest that adding or repairing modalities may not always improve prediction, and that the utility of each modality is sample-dependent. Building on this finding, we propose \textbf{S}ufficiency-\textbf{I}nformed \textbf{E}vidential \textbf{V}al\textbf{vE} (\textbf{SIEVE}) that turns ``whether to repair'' into an explicit, learnable decision at the sample level. SIEVE compares a direct prediction branch with a repair branch, derives an empirical sufficiency signal from their per-sample loss gap, and routes each input through an evidential gate that jointly models sufficiency and its epistemic uncertainty. SIEVE is repair-agnostic: it operates as a plug-and-play decision on top of any explicit or implicit repair module, without modifying its internal design. Experiments on CMU-MOSI and IEMOCAP show that SIEVE consistently improves representative repair backbones across evaluated missing rates, and approaches the per-sample dual-branch achievable optimum.
