Yes FLoReNce, I Will Do Better Next Time! Agentic Feedback Reasoning for Humorous Meme Detection

2026-01-12

AI Summary

This work addresses the challenge that existing models struggle to capture the satirical and socio-critical intent of humorous memes and lack the ability to self-correct during reasoning. To this end, the authors propose FLoReNce, a novel framework that introduces, for the first time, an agent-based feedback-driven closed-loop learning mechanism. During training, an agent interacts with a critic module to generate semantic feedback, which is stored in a non-parametric knowledge bank. At inference, multimodal prompts are combined with retrieved knowledge from this bank, and semantic control signals modulate the reasoning process, enabling adaptive optimization without fine-tuning. Experiments on the PrideMM dataset demonstrate significant improvements in both prediction accuracy and explanation quality, validating the effectiveness of feedback-modulated prompting for understanding humorous memes.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Multiagent Systems: Multiagent Learning
• Natural Language Processing: Prompt Engineering / Prompting

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Humorous memes blend visual and textual cues to convey irony, satire, or social commentary, posing unique challenges for AI systems that must interpret intent rather than surface correlations. Existing multimodal or prompting-based models generate explanations for humor but operate in an open loop,lacking the ability to critique or refine their reasoning once a prediction is made. We propose FLoReNce, an agentic feedback reasoning framework that treats meme understanding as a closed-loop process during learning and an open-loop process during inference. In the closed loop, a reasoning agent is critiqued by a judge; the error and semantic feedback are converted into control signals and stored in a feedback-informed, non-parametric knowledge base. At inference, the model retrieves similar judged experiences from this KB and uses them to modulate its prompt, enabling better, self-aligned reasoning without finetuning. On the PrideMM dataset, FLoReNce improves both predictive performance and explanation quality over static multimodal baselines, showing that feedback-regulated prompting is a viable path to adaptive meme humor understanding.

Problem

• humorous meme detection
• multimodal reasoning
• feedback reasoning
• intent interpretation
• explanation refinement

Innovation

• agentic feedback reasoning
• closed-loop learning
• feedback-informed knowledge base
• adaptive prompting
• humorous meme detection