# Segregate, Refine, Integrate: Decomposing Multimodal Fusion for Sentiment Analysis

arXiv: https://arxiv.org/abs/2607.12686 · HTML: https://mlsanity.com/paper/2607.12686/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 39 references that are in the map.
- First posted on arXiv 14 Jul 2026.

## Builds on (most cited first, 19 of 19)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Flamingo: a Visual Language Model for Few-Shot Learning](https://mlsanity.com/paper/2204.14198/) (2022)
- [Multimodal Machine Learning: A Survey and Taxonomy](https://mlsanity.com/paper/1705.09406/) (2017)
- [Multimodal Transformer for Unaligned Multimodal Language Sequences](https://mlsanity.com/paper/1906.00295/) (2019)
- [Tensor Fusion Network for Multimodal Sentiment Analysis](https://mlsanity.com/paper/1707.07250/) (2017)
- [Generating Wikipedia by Summarizing Long Sequences](https://mlsanity.com/paper/1801.10198/) (2018)
- [Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free](https://mlsanity.com/paper/2505.06708/) (2025)
- [Efficient Low-rank Multimodal Fusion with Modality-Specific Factors](https://mlsanity.com/paper/1806.00064/) (2018)
- [MISA: Modality-Invariant and -Specific Representations for Multimodal Sentiment Analysis](https://mlsanity.com/paper/2005.03545/) (2020)
- [Audio Flamingo: A Novel Audio Language Model with Few-Shot Learning and Dialogue Abilities](https://mlsanity.com/paper/2402.01831/) (2024)
- [Learning Modality-Specific Representations with Self-Supervised Multi-Task Learning for Multimodal Sentiment Analysis](https://mlsanity.com/paper/2102.04830/) (2021)
- [UER: An Open-Source Toolkit for Pre-training Models](https://mlsanity.com/paper/1909.05658/) (2019)
- [An Embarrassingly Simple Approach for Transfer Learning from Pretrained Language Models](https://mlsanity.com/paper/1902.10547/) (2019)
- [DLF: Disentangled-Language-Focused Multimodal Sentiment Analysis](https://mlsanity.com/paper/2412.12225/) (2024)
- [Enriching Multimodal Sentiment Analysis through Textual Emotional Descriptions of Visual-Audio Content](https://mlsanity.com/paper/2412.10460/) (2024)
- [M-SENA: An Integrated Platform for Multimodal Sentiment Analysis](https://mlsanity.com/paper/2203.12441/) (2022)
- [Knowledge-Guided Dynamic Modality Attention Fusion Framework for Multimodal Sentiment Analysis](https://mlsanity.com/paper/2410.04491/) (2024)
- [SeqAug: Sequential Feature Resampling as a modality agnostic augmentation method](https://mlsanity.com/paper/2305.01954/) (2023)
- [DeepMLF: Multimodal language model with learnable tokens for deep fusion in sentiment analysis](https://mlsanity.com/paper/2504.11082/) (2025)

## Abstract (arXiv)

Multimodal fusion must simultaneously refine modality-specific signals and model cross-modal interactions; two competing objectives typically entangled within the same operation. We propose SeRIn (Segregate, Refine, Integrate), a multimodal LM fusion scheme that enforces this separation as an architectural prior. Modality-specific representations evolve along isolated pathways, each refined against its respective encoder context, while a dedicated cross-modal pathway accumulates their joint evolution without contaminating unimodal streams. Full cross-modal interaction is deferred to a final prediction step - ablations confirm that structured interactions, not added capacity, drive the gains; gate analysis under visual corruption reveals emergent modality reweighting without explicit supervision. SeRIn achieves state-of-the-art results on CH-SIMS and CMU-MOSEI, improving all metrics on both benchmarks.