# C2F-Thinker: Coarse-to-Fine Reasoning with Hint-Guided Reinforcement Learning for Multimodal Sentiment Analysis

arXiv: https://arxiv.org/abs/2604.00013 · HTML: https://mlsanity.com/paper/2604.00013/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Audio-language model reasoning, one of 17 directions in LLM reasoning.
- Builds on 18 of its 31 references that are in the map.
- First posted on arXiv 10 Mar 2026.

## Placed in

- LLM reasoning / [Audio-language model reasoning](https://mlsanity.com/lr/audio-language-model-reasoning/)

## Builds on (most cited first, 18 of 18)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding](https://mlsanity.com/paper/2306.02858/) (2023)
- [MiniCPM-V: A GPT-4V Level MLLM on Your Phone](https://mlsanity.com/paper/2408.01800/) (2024)
- [Multimodal Chain-of-Thought Reasoning in Language Models](https://mlsanity.com/paper/2302.00923/) (2023)
- [Qwen2.5-Omni Technical Report](https://mlsanity.com/paper/2503.20215/) (2025)
- [Qwen3-Omni Technical Report](https://mlsanity.com/paper/2509.17765/) (2025)
- [PandaGPT: One Model To Instruction-Follow Them All](https://mlsanity.com/paper/2305.16355/) (2023)
- [Reinforcement Learning with Verifiable Rewards Implicitly Incentivizes Correct Reasoning in Base LLMs](https://mlsanity.com/paper/2506.14245/) (2025)
- [Multimodal Chain-of-Thought Reasoning: A Comprehensive Survey](https://mlsanity.com/paper/2503.12605/) (2025)
- [HumanOmni: A Large Vision-Speech Language Model for Human-Centric Video Understanding](https://mlsanity.com/paper/2501.15111/) (2025)
- [Think or Not Think: A Study of Explicit Thinking in Rule-Based Visual Reinforcement Fine-Tuning](https://mlsanity.com/paper/2503.16188/) (2025)
- [Multimodal Large Language Models Meet Multimodal Emotion Recognition and Reasoning: A Survey](https://mlsanity.com/paper/2509.24322/) (2025)
- [AffectGPT-R1: Leveraging Reinforcement Learning for Open-Vocabulary Multimodal Emotion Recognition](https://mlsanity.com/paper/2508.01318/) (2025)
- [Can Large Language Models Help Multimodal Language Analysis? MMLA: A Comprehensive Benchmark](https://mlsanity.com/paper/2504.16427/) (2025)
- [GenCLS++: Pushing the Boundaries of Generative Classification in LLMs Through Comprehensive SFT and RL Studies Across Diverse Datasets](https://mlsanity.com/paper/2504.19898/) (2025)
- [Multimodal Large Language Models for End-to-End Affective Computing: Benchmarking and Boosting with Generative Knowledge Prompting](https://mlsanity.com/paper/2508.02429/) (2025)
- [Exploring Large Language Models for Multimodal Sentiment Analysis: Challenges, Benchmarks, and Future Directions](https://mlsanity.com/paper/2411.15408/) (2024)

## Abstract (arXiv)

Multimodal sentiment analysis aims to integrate textual, acoustic, and visual information for deep emotional understanding. Despite the progress of multimodal large language models (MLLMs) via supervised fine-tuning, their "black-box" nature hinders interpretability. While Chain-of-Thought (CoT) reasoning offers a potential remedy, it is constrained by high manual annotation costs and the inherent challenges of reinforcement learning (RL), such as reward sparsity and low exploration efficiency on hard samples. This paper presents C2F-Thinker, a framework that harmonizes coarse-to-fine structured reasoning with hint-guided RL through a two-stage progressive training pipeline. In the first stage, we conduct cold-start supervised fine-tuning using high-quality CoT data distilled from a larger teacher model, consisting of three distinct phases: polarity judgment, intermediate analysis, and fine-grained scoring. This equips the base model with a structured emotional reasoning paradigm. In the second stage, we introduce a hint-guided Group Relative Policy Optimization (GRPO) algorithm. By injecting correct initial polarity predictions as hints during the sampling process, the model is guided toward accurate reasoning paths, effectively mitigating cascading errors and enhancing the utilization of hard samples. Furthermore, a multi-faceted reward function incorporating classification, regression, and formatting constraints is designed to refine prediction accuracy while preserving interpretability. Experimental results demonstrate that C2F-Thinker achieves competitive performance on fine-grained sentiment regression tasks while significantly outperforming baselines in cross-domain generalization. This highlights its potential in building trustworthy and robust sentiment analysis systems for real-world applications.