TinyEmo：通過度量投影縮小情感推理

TinyEmo: Scaling down Emotional Reasoning via Metric Projection

October 9, 2024

作者: Cristian Gutierrez

cs.AI

摘要

本文介紹了TinyEmo，這是一系列用於情感推理和分類的小型多模態語言模型。我們的方法包括：(1) 一個用於預訓練和微調階段的合成情感指導數據集，(2) 一個度量投影器，將分類從語言模型中分離出來，從而實現更高效的訓練和推理，(3) 一個用於情感推理的多模態大型語言模型（MM-LLM），以及(4) 一個半自動化的偏見檢測框架。TinyEmo能夠執行情感分類和情感推理，同時使用的參數比可比模型少得多。這種高效性使我們能夠自由地整合更多不同的情感數據集，在分類任務上實現出色的性能，我們最小的模型（700M參數）在基於通用MM-LLM且具有超過7B參數的更大型最新模型上表現更好。此外，度量投影器允許在大型模型中進行可解釋性和間接偏見檢測，無需額外的訓練，提供了一種理解和改進AI系統的方法。 我們在https://github.com/ggcr/TinyEmo 上發布了代碼、模型和數據集。

English

This paper introduces TinyEmo, a family of small multi-modal language models for emotional reasoning and classification. Our approach features: (1) a synthetic emotional instruct dataset for both pre-training and fine-tuning stages, (2) a Metric Projector that delegates classification from the language model allowing for more efficient training and inference, (3) a multi-modal large language model (MM-LLM) for emotional reasoning, and (4) a semi-automated framework for bias detection. TinyEmo is able to perform emotion classification and emotional reasoning, all while using substantially fewer parameters than comparable models. This efficiency allows us to freely incorporate more diverse emotional datasets, enabling strong performance on classification tasks, with our smallest model (700M parameters) outperforming larger state-of-the-art models based on general-purpose MM-LLMs with over 7B parameters. Additionally, the Metric Projector allows for interpretability and indirect bias detection in large models without additional training, offering an approach to understand and improve AI systems. We release code, models, and dataset at https://github.com/ggcr/TinyEmo