CMHL: Contrastive Multi-Head Learning for Emotionally Consistent Text Classification

TL;DR

This paper introduces CMHL, a compact single-model architecture that enhances emotion classification accuracy and consistency by integrating psychological models and contrastive loss, outperforming larger models and ensembles.

Contribution

The paper presents a novel single-model approach with multi-task learning, psychological supervision, and a contrastive contradiction loss for emotionally consistent text classification.

Findings

Outperforms larger LLMs and ensembles with 125M parameters.

Achieves state-of-the-art F1 score of 93.75% on the dair-ai dataset.

Demonstrates strong cross-domain generalization and improved mental health detection.

Abstract

Textual Emotion Classification (TEC) is one of the most difficult NLP tasks. State of the art approaches rely on Large language models (LLMs) and multi-model ensembles. In this study, we challenge the assumption that larger scale or more complex models are necessary for improved performance. In order to improve logical consistency, We introduce CMHL, a novel single-model architecture that explicitly models the logical structure of emotions through three key innovations: (1) multi-task learning that jointly predicts primary emotions, valence, and intensity, (2) psychologically-grounded auxiliary supervision derived from Russell's circumplex model, and (3) a novel contrastive contradiction loss that enforces emotional consistency by penalizing mutually incompatible predictions (e.g., simultaneous high confidence in joy and anger). With just 125M parameters, our model outperforms 56x larger LLMs and sLM ensembles with a new state-of-the-art F1 score of 93.75\% compared to (86.13\%-93.2\%) on the dair-ai Emotion dataset. We further show cross domain generalization on the Reddit Suicide Watch and Mental Health Collection dataset (SWMH), outperforming domain-specific models like MentalBERT and MentalRoBERTa with an F1 score of 72.50\% compared to (68.16\%-72.16\%) + a 73.30\% recall compared to (67.05\%-70.89\%) that translates to enhanced sensitivity for detecting mental health distress. Our work establishes that architectural intelligence (not parameter count) drives progress in TEC. By embedding psychological priors and explicit consistency constraints, a well-designed single model can outperform both massive LLMs and complex ensembles, offering a efficient, interpretable, and clinically-relevant paradigm for affective computing.