LLaMAs Have Feelings Too: Unveiling Sentiment and Emotion Representations in LLaMA Models Through Probing

• URL: http://arxiv.org/abs/2505.16491v2
• Date: Fri, 30 May 2025 10:15:03 GMT
• Title: LLaMAs Have Feelings Too: Unveiling Sentiment and Emotion Representations in LLaMA Models Through Probing
• Authors: Dario Di Palma, Alessandro De Bellis, Giovanni Servedio, Vito Walter Anelli, Fedelucio Narducci, Tommaso Di Noia,
• Abstract summary: This study probes the hidden layers of Llama models to pinpoint where sentiment features are most represented.<n>We analyze sentiment encoding across layers and scales, identifying the layers and pooling methods that best capture sentiment signals.
• Score: 47.20927495079714
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) have rapidly become central to NLP, demonstrating their ability to adapt to various tasks through prompting techniques, including sentiment analysis. However, we still have a limited understanding of how these models capture sentiment-related information. This study probes the hidden layers of Llama models to pinpoint where sentiment features are most represented and to assess how this affects sentiment analysis. Using probe classifiers, we analyze sentiment encoding across layers and scales, identifying the layers and pooling methods that best capture sentiment signals. Our results show that sentiment information is most concentrated in mid-layers for binary polarity tasks, with detection accuracy increasing up to 14% over prompting techniques. Additionally, we find that in decoder-only models, the last token is not consistently the most informative for sentiment encoding. Finally, this approach enables sentiment tasks to be performed with memory requirements reduced by an average of 57%. These insights contribute to a broader understanding of sentiment in LLMs, suggesting layer-specific probing as an effective approach for sentiment tasks beyond prompting, with potential to enhance model utility and reduce memory requirements.

Related papers

• Spectral Insights into Data-Oblivious Critical Layers in Large Language Models [7.486925126518052]
  We introduce a data-oblivious approach to identify intrinsic critical layers in pre-fine-tuned language models.<n>We show that layers with significant shifts in representation space are also those most affected during fine-tuning.
  arXiv  Detail & Related papers  (2025-05-31T04:21:39Z)
• DeepPerception: Advancing R1-like Cognitive Visual Perception in MLLMs for Knowledge-Intensive Visual Grounding [61.26026947423187]
  Human experts excel at fine-grained visual discrimination by leveraging domain knowledge to refine perceptual features.<n>Current Multimodal Large Language Models (MLLMs) struggle to integrate reasoning into visual perception.<n>We propose DeepPerception, an MLLM enhanced with cognitive visual perception capabilities.
  arXiv  Detail & Related papers  (2025-03-17T04:06:34Z)
• Integration of Explainable AI Techniques with Large Language Models for Enhanced Interpretability for Sentiment Analysis [0.5120567378386615]
  Interpretability remains a key difficulty in sentiment analysis with Large Language Models (LLMs)<n>This research introduces a technique that applies SHAP (Shapley Additive Explanations) by breaking down LLMs into components such as embedding layer,encoder,decoder and attention layer.<n>The method is evaluated using the Stanford Sentiment Treebank (SST-2) dataset, which shows how different sentences affect different layers.
  arXiv  Detail & Related papers  (2025-03-15T01:37:54Z)
• Human Cognitive Benchmarks Reveal Foundational Visual Gaps in MLLMs [65.93003087656754]
  VisFactor is a benchmark that digitizes 20 vision-centric subtests from a well-established cognitive psychology assessment.<n>We evaluate 20 frontier Multimodal Large Language Models (MLLMs) from GPT, Gemini, Claude, LLaMA, Qwen, and SEED families.<n>The best-performing model achieves a score of only 25.19 out of 100, with consistent failures on tasks such as mental rotation, spatial relation inference, and figure-ground discrimination.
  arXiv  Detail & Related papers  (2025-02-23T04:21:32Z)
• Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding [58.364933651703524]
  We show that concentrated massive values consistently emerge in specific regions of attention queries.<n>These massive values play a critical role in interpreting contextual knowledge.<n>We trace the emergence of massive values and find that such concentration is caused by Rotary Positional.
  arXiv  Detail & Related papers  (2025-02-03T17:47:03Z)
• Do Large Language Models Possess Sensitive to Sentiment? [18.88126980975737]
  Large Language Models (LLMs) have recently displayed their extraordinary capabilities in language understanding.<n>This paper investigates the ability of LLMs to detect and react to sentiment in text modal.
  arXiv  Detail & Related papers  (2024-09-04T01:40:20Z)
• EmoLLMs: A Series of Emotional Large Language Models and Annotation Tools for Comprehensive Affective Analysis [33.0280076936761]
  We propose EmoLLMs, the first series of open-sourced instruction-following LLMs for affective analysis based on fine-tuning various LLMs with instruction data. EmoLLMs achieve the ChatGPT-level and GPT-4-level generalization capabilities on affective analysis tasks, and demonstrates our models can be used as affective annotation tools.
  arXiv  Detail & Related papers  (2024-01-16T17:11:11Z)
• Deep Imbalanced Learning for Multimodal Emotion Recognition in Conversations [15.705757672984662]
  Multimodal Emotion Recognition in Conversations (MERC) is a significant development direction for machine intelligence. Many data in MERC naturally exhibit an imbalanced distribution of emotion categories, and researchers ignore the negative impact of imbalanced data on emotion recognition. We propose the Class Boundary Enhanced Representation Learning (CBERL) model to address the imbalanced distribution of emotion categories in raw data. We have conducted extensive experiments on the IEMOCAP and MELD benchmark datasets, and the results show that CBERL has achieved a certain performance improvement in the effectiveness of emotion recognition.
  arXiv  Detail & Related papers  (2023-12-11T12:35:17Z)
• How are Prompts Different in Terms of Sensitivity? [50.67313477651395]
  We present a comprehensive prompt analysis based on the sensitivity of a function. We use gradient-based saliency scores to empirically demonstrate how different prompts affect the relevance of input tokens to the output. We introduce sensitivity-aware decoding which incorporates sensitivity estimation as a penalty term in the standard greedy decoding.
  arXiv  Detail & Related papers  (2023-11-13T10:52:01Z)
• MMLatch: Bottom-up Top-down Fusion for Multimodal Sentiment Analysis [84.7287684402508]
  Current deep learning approaches for multimodal fusion rely on bottom-up fusion of high and mid-level latent modality representations. Models of human perception highlight the importance of top-down fusion, where high-level representations affect the way sensory inputs are perceived. We propose a neural architecture that captures top-down cross-modal interactions, using a feedback mechanism in the forward pass during network training.
  arXiv  Detail & Related papers  (2022-01-24T17:48:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.