# Group Cognition Learning: Making Everything Better Through Governed Two-Stage Agents Collaboration

arXiv: https://arxiv.org/abs/2605.00370 · HTML: https://mlsanity.com/paper/2605.00370/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 28 references that are in the map.
- First posted on arXiv 1 May 2026.

## Builds on (most cited first, 17 of 17)

- [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)
- [Efficient Low-rank Multimodal Fusion with Modality-Specific Factors](https://mlsanity.com/paper/1806.00064/) (2018)
- [Learning Modality-Specific Representations with Self-Supervised Multi-Task Learning for Multimodal Sentiment Analysis](https://mlsanity.com/paper/2102.04830/) (2021)
- [MIntRec: A New Dataset for Multimodal Intent Recognition](https://mlsanity.com/paper/2209.04355/) (2022)
- [Classifier-guided Gradient Modulation for Enhanced Multimodal Learning](https://mlsanity.com/paper/2411.01409/) (2024)
- [Towards Multimodal Sentiment Analysis Debiasing via Bias Purification](https://mlsanity.com/paper/2403.05023/) (2024)
- [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)
- [Boosting Multimodal Learning via Disentangled Gradient Learning](https://mlsanity.com/paper/2507.10213/) (2025)
- [Fine-grained Disentangled Representation Learning for Multimodal Emotion Recognition](https://mlsanity.com/paper/2312.13567/) (2023)
- [Tri-Subspaces Disentanglement for Multimodal Sentiment Analysis](https://mlsanity.com/paper/2602.19585/) (2026)
- [Multimodal Transformers are Hierarchical Modal-wise Heterogeneous Graphs](https://mlsanity.com/paper/2505.01068/) (2025)
- [CLCR: Cross-Level Semantic Collaborative Representation for Multimodal Learning](https://mlsanity.com/paper/2602.19605/) (2026)
- [MInD: Improving Multimodal Sentiment Analysis via Multimodal Information Disentanglement](https://mlsanity.com/paper/2401.11818/) (2024)
- [Temporal-Spatial Decouple before Act: Disentangled Representation Learning for Multimodal Sentiment Analysis](https://mlsanity.com/paper/2601.13659/) (2026)
- [Mitigating Shared-Private Branch Imbalance via Dual-Branch Rebalancing for Multimodal Sentiment Analysis](https://mlsanity.com/paper/2604.25179/) (2026)

## Built on by (most cited first, 3 of 3)

- [Rethinking Modality Reliability in Multimodal Sentiment Analysis with Incomplete Observations](https://mlsanity.com/paper/2608.03611/) (2026)
- [CM2: Multimodal Cultural Reasoning via an Integrated Multi-Agent Framework](https://mlsanity.com/paper/2608.30498/) (2026)
- [SAVOR: Self-Aware Visual Grounding via Confidence-Calibrated Reinforcement Learning for Multimodal Hallucination Mitigation](https://mlsanity.com/paper/2609.16601/) (2026)

## Abstract (arXiv)

Centralized multimodal learning commonly compresses language, acoustic, and visual signals into a single fused representation for prediction. While effective, this paradigm suffers from two limitations: modality dominance, where optimization gravitates towards the path of least resistance, ignoring weaker but informative modalities, and spurious modality coupling, where models overfit to incidental cross-modal correlations. To address these, we propose Group Cognition Learning (GCL), a governed collaboration paradigm that applies a two-stage protocol after modality-specific encoding. In Stage 1 (Selective Interaction), a Routing Agent proposes directed interaction routes, and an Auditing Agent assigns sample-wise gates to emphasize exchanges that yield positive marginal predictive gain while suppressing redundant coupling. In Stage 2 (Consensus Formation), a Public-Factor Agent maintains an explicit shared factor, and an Aggregation Agent produces the final prediction through contribution-aware weighting while keeping each modality representation as a specialization channel. Extensive experiments on CMU-MOSI, CMU-MOSEI, and MIntRec demonstrate that GCL mitigates dominance and coupling, establishing state-of-the-art results across both regression and classification benchmarks. Analysis experiments further demonstrate the effectiveness of the design.