Search papers, labs, and topics across Lattice.
This paper introduces the Modality Reliability-Calibrated Framework (MRCF) for Multimodal Sentiment Analysis (MSA) that explicitly models modality reliability in scenarios with incomplete observations. By incorporating a Reliability-Aware Branch, a Reliability-Guided Interaction Branch, and a Reliability-Calibrated Fusion Module, MRCF addresses issues of reliability mismatch and propagation bias that typically hinder predictive performance in existing methods. Experimental results on multiple datasets demonstrate that MRCF significantly improves sentiment analysis accuracy by effectively leveraging sample-specific modality reliability.
Explicitly modeling modality reliability can drastically enhance sentiment analysis performance in the face of incomplete multimodal data.
Multimodal Sentiment Analysis (MSA) integrates text, audio, and vision to infer human affect, yet real-world multimodal observations are often incomplete. Existing methods for incomplete-observation MSA mainly follow two paradigms. Reconstruction-based methods recover missing information from observed modalities, while joint-representation methods learn directly from incomplete inputs. Although effective, these methods usually treat modality reliability only implicitly within representation learning or fusion design rather than modeling it explicitly. We argue that modality reliability is a central variable in incomplete-observation settings. Failure to model it explicitly gives rise to two related issues. The first is reliability mismatch, in which the affective evidence retained by each modality varies across samples and missing rates. The second is reliability propagation bias, in which messages from degraded modalities may adversely affect cross-modal interaction and predictive performance. To address these issues, we propose MRCF, a Modality Reliability-Calibrated Framework for MSA with incomplete observations. MRCF contains a Reliability-Aware Branch that estimates sample-specific modality reliability from intramodal quality cues and cross-modal semantic consistency, a Reliability-Guided Interaction Branch that uses the estimated scores to modulate cross-modal information flow, and a Reliability-Calibrated Fusion Module that integrates reliability and semantic cues for final prediction. Experiments on CMU-MOSI, CMU-MOSEI, and CH-SIMS show that MRCF achieves strong performance under standard incomplete-observation protocols. Further analyses provide evidence that explicit reliability modeling helps mitigate reliability mismatch and reliability propagation bias during interaction and fusion.
