Fine-Grained Emotion Recognition via In-Context Learning

2025-10-07

AI Summary

Existing in-context learning (ICL) methods for fine-grained sentiment recognition emphasize reasoning while neglecting decision-making mechanisms, rendering them vulnerable to bias from semantically similar but sentiment-divergent exemplars—thereby compromising sentiment representation accuracy. Grounded in prototype theory, this work identifies decision fragility in the matching process between query representations and sentiment prototypes. We propose Emotion-aware In-Context Learning (EICL), which (1) constructs an emotion-similarity-based exemplar retrieval mechanism, (2) introduces dynamic soft labeling to model sentiment uncertainty, and (3) designs a two-stage exclusion strategy—semantic-emotion dual filtering—to refine exemplar selection. Evaluated across multiple fine-grained sentiment datasets, EICL consistently outperforms standard ICL, achieving average accuracy gains of 3.2–5.7 percentage points. To our knowledge, this is the first work to systematically integrate prototype-driven decision modeling into the ICL framework, establishing a novel paradigm for interpretable and robust sentiment recognition.

Technology Category

• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Machine Learning: Learning Preferences or Rankings
• Knowledge Representation and Reasoning: Preferences

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Fine-grained emotion recognition aims to identify the emotional type in queries through reasoning and decision-making processes, playing a crucial role in various systems. Recent methods use In-Context Learning (ICL), enhancing the representation of queries in the reasoning process through semantically similar examples, while further improving emotion recognition by explaining the reasoning mechanisms. However, these methods enhance the reasoning process but overlook the decision-making process. This paper investigates decision-making in fine-grained emotion recognition through prototype theory. We show that ICL relies on similarity matching between query representations and emotional prototypes within the model, where emotion-accurate representations are critical. However, semantically similar examples often introduce emotional discrepancies, hindering accurate representations and causing errors. To address this, we propose Emotion In-Context Learning (EICL), which introduces emotionally similar examples and uses a dynamic soft-label strategy to improve query representations in the emotion reasoning process. A two-stage exclusion strategy is then employed to assess similarity from multiple angles, further optimizing the decision-making process. Extensive experiments show that EICL significantly outperforms ICL on multiple datasets.

Problem

• Fine-grained emotion recognition overlooks decision-making process in ICL methods
• Semantically similar examples introduce emotional discrepancies causing representation errors
• Proposes EICL with emotional examples and two-stage exclusion for optimization

Innovation

• Uses emotionally similar examples for representation
• Applies dynamic soft-label strategy in reasoning
• Implements two-stage exclusion for decision optimization