# From Perception to Punchline: Empowering VLM with the Art of In-the-wild Meme

arXiv: https://arxiv.org/abs/2512.24555 · HTML: https://mlsanity.com/paper/2512.24555/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 38 of its 61 references that are in the map.
- First posted on arXiv 31 Dec 2025.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)

## Builds on (most cited first, 30 of 38)

- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models](https://mlsanity.com/paper/2504.10479/) (2025)
- [Multimodal Chain-of-Thought Reasoning in Language Models](https://mlsanity.com/paper/2302.00923/) (2023)
- [Show-o: One Single Transformer to Unify Multimodal Understanding and Generation](https://mlsanity.com/paper/2408.12528/) (2024)
- [HuatuoGPT-o1, Towards Medical Complex Reasoning with LLMs](https://mlsanity.com/paper/2412.18925/) (2024)
- [Beyond One-Preference-Fits-All Alignment: Multi-Objective Direct Preference Optimization](https://mlsanity.com/paper/2310.03708/) (2023)
- [Modular Pluralism: Pluralistic Alignment via Multi-LLM Collaboration](https://mlsanity.com/paper/2406.15951/) (2024)
- [The CoT Collection: Improving Zero-shot and Few-shot Learning of Language Models via Chain-of-Thought Fine-Tuning](https://mlsanity.com/paper/2305.14045/) (2023)
- [Do Androids Laugh at Electric Sheep? Humor "Understanding" Benchmarks from The New Yorker Caption Contest](https://mlsanity.com/paper/2209.06293/) (2022)
- [SemEval-2020 Task 8: Memotion Analysis -- The Visuo-Lingual Metaphor!](https://mlsanity.com/paper/2008.03781/) (2020)
- [Natural Language Understanding and Inference with MLLM in Visual Question Answering: A Survey](https://mlsanity.com/paper/2411.17558/) (2024)
- [Visual Program Distillation: Distilling Tools and Programmatic Reasoning into Vision-Language Models](https://mlsanity.com/paper/2312.03052/) (2023)
- [Kwai Keye-VL Technical Report](https://mlsanity.com/paper/2507.01949/) (2025)
- [MemeCap: A Dataset for Captioning and Interpreting Memes](https://mlsanity.com/paper/2305.13703/) (2023)
- [MonoFormer: One Transformer for Both Diffusion and Autoregression](https://mlsanity.com/paper/2409.16280/) (2024)
- [Deep Learning Approaches on Image Captioning: A Review](https://mlsanity.com/paper/2201.12944/) (2022)
- [Unifying Multimodal Transformer for Bi-directional Image and Text Generation](https://mlsanity.com/paper/2110.09753/) (2021)
- [ERNIE-ViLG: Unified Generative Pre-training for Bidirectional Vision-Language Generation](https://mlsanity.com/paper/2112.15283/) (2021)
- [From Local Concepts to Universals: Evaluating the Multicultural Understanding of Vision-Language Models](https://mlsanity.com/paper/2407.00263/) (2024)
- [XMeCap: Meme Caption Generation with Sub-Image Adaptability](https://mlsanity.com/paper/2407.17152/) (2024)
- [Beyond Words: Exploring Cultural Value Sensitivity in Multimodal Models](https://mlsanity.com/paper/2502.14906/) (2025)
- [MEMEX: Detecting Explanatory Evidence for Memes via Knowledge-Enriched Contextualization](https://mlsanity.com/paper/2305.15913/) (2023)
- [Dank Learning: Generating Memes Using Deep Neural Networks](https://mlsanity.com/paper/1806.04510/) (2018)

## Built on by (most cited first, 1 of 1)

- [MER-Bench: A Comprehensive Benchmark for Multimodal Meme Reappraisal](https://mlsanity.com/paper/2603.15020/) (2026)

## Abstract (arXiv)

Generating humorous memes is a challenging multimodal task that moves beyond direct image-to-caption supervision. It requires a nuanced reasoning over visual content, contextual cues, and subjective humor. To bridge this gap between visual perception and humorous punchline creation, we propose HUMOR}, a novel framework that guides VLMs through hierarchical reasoning and aligns them with group-wise human preferences. First, HUMOR employs a hierarchical, multi-path Chain-of-Thought (CoT): the model begins by identifying a template-level intent, then explores diverse reasoning paths under different contexts, and finally anchors onto a high-quality, context-specific path. This CoT supervision, which traces back from ground-truth captions, enhances reasoning diversity. We further analyze that this multi-path exploration with anchoring maintains a high expected humor quality, under the practical condition that high-quality paths retain significant probability mass. Second, to capture subjective humor, we train a pairwise reward model that operates within groups of memes sharing the same template. Following established theory, this approach ensures a consistent and robust proxy for human preference, even with subjective and noisy labels. The reward model then enables a group-wise reinforcement learning optimization, guaranteeing providing a theoretical guarantee for monotonic improvement within the trust region. Extensive experiments show that HUMOR empowers various VLMs with superior reasoning diversity, more reliable preference alignment, and higher overall meme quality. Beyond memes, our work presents a general training paradigm for open-ended, human-aligned multimodal generation, where success is guided by comparative judgment within coherent output group.