analiza aiqbench
This research addresses the deployment of multimodal sentiment analysis in resource-constrained settings by introducing a lightweight model that generates reasoning chains alongside classification results. Using a "Teacher-Assistant-Student" distillation framework, the proposed method achieves strong performance with only 3B parameters while enhancing interpretability without relying on massive LLMs during inference.
Resource-Limited Joint Multimodal Sentiment Reasoning and Classification via Chain-of-Thought Enhancement and Distillation
arXiv:2508.05234v2 Announce Type: replace-cross Abstract: Current approaches for Multimodal Sentiment Analysis (MSA) primarily leverage the knowledge and reasoning capabilities of parameter-heavy (Multimodal) LLMs for classification, overlooking autonomous multimodal sentiment reasoning generation in resource-constrained environments. In this paper, we focus on the Resource-Limited Joint Multimodal Sentiment Reasoning and Classification task, JMSRC, which simultaneously performs multimodal sentiment reasoning chain generation and sentiment classification only with a lightweight model. We propose a Multimodal Chain-of-Thought Reasoning Distillation model, MulCoT-RD, designed for JMSRC that employs a "Teacher-Assistant-Student" distillation paradigm to address deployment constraints in resource-limited environments. We first leverage a high-performance Multimodal Large Language Model (MLLM) to generate the initial reasoning dataset and train a medium-sized assistant model with a multi-task learning mechanism. A lightweight student model is jointly trained to perform efficient multimodal sentiment reasoning generation and classification. Extensive experiments on four datasets demonstrate that MulCoT-RD, with only 3B parameters, achieves strong performance on JMSRC while exhibiting robust generalization and enhanced interpretability.
Abstract:Current approaches for Multimodal Sentiment Analysis (MSA) primarily leverage the knowledge and reasoning capabilities of parameter-heavy (Multimodal) LLMs for classification, overlooking autonomous multimodal sentiment reasoning generation in resource-constrained environments. In this paper, we focus on the Resource-Limited Joint Multimodal Sentiment Reasoning and Classification task, JMSRC, which simultaneously performs multimodal sentiment reasoning chain generation and sentiment classification only with a lightweight model. We propose a Multimodal Chain-of-Thought Reasoning Distillation model, MulCoT-RD, designed for JMSRC that employs a "Teacher-Assistant-Student" distillation paradigm to address deployment constraints in resource-limited environments. We first leverage a high-performance Multimodal Large Language Model (MLLM) to generate the initial reasoning dataset and train a medium-sized assistant model with a multi-task learning mechanism. A lightweight student model is jointly trained to perform efficient multimodal sentiment reasoning generation and classification. Extensive experiments on four datasets demonstrate that MulCoT-RD, with only 3B parameters, achieves strong performance on JMSRC while exhibiting robust generalization and enhanced interpretability.
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
arXiv:2508.05234 [cs.CL]
(or arXiv:2508.05234v2 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2508.05234 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 249 cuvinte
Subiecte conexe
LiveLaw Supreme Court Half-Yearly Digest: January - June 2026 [Part 2]
livelaw.in · 13 sept. 2026
GRASP: Grounded CoT Reasoning with Dual-Stage Optimization for Multimodal Sarcasm Target Identification
arXiv · 21 iul. 2026
A Red Line and Oversight Framework for Government AI Contracts
AI Alignment Forum · 18 iul. 2026
Nemotron 3.5 Content Safety: Customizable Multimodal Safety for Global Enterprise AI
Hugging Face · 4 iun. 2026