# MMDS-Bench: Benchmarking Multimodal Large Language Models on Dynamic Stance in Social Media Interactions

arXiv: https://arxiv.org/abs/2608.30903 · HTML: https://mlsanity.com/paper/2608.30903/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in MLLM training and visual instruction tuning, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 19 of its 47 references that are in the map.
- First posted on arXiv 31 Aug 2026.

## Placed in

- Multimodal LLMs / vision-language models / [MLLM training and visual instruction tuning](https://mlsanity.com/mllm/mllm-training-and-visual-instruction-tuning/)

## Builds on (most cited first, 19 of 19)

- [MMBench: Is Your Multi-modal Model an All-around Player?](https://mlsanity.com/paper/2307.06281/) (2023)
- [MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilities](https://mlsanity.com/paper/2308.02490/) (2023)
- [Stance Detection on Social Media: State of the Art and Trends](https://mlsanity.com/paper/2006.03644/) (2020)
- [Stance Detection with Bidirectional Conditional Encoding](https://mlsanity.com/paper/1606.05464/) (2016)
- [Cross-Target Stance Classification with Self-Attention Networks](https://mlsanity.com/paper/1805.06593/) (2018)
- [Multi-modal Stance Detection: New Datasets and Model](https://mlsanity.com/paper/2402.14298/) (2024)
- [A Challenge Dataset and Effective Models for Conversational Stance Detection](https://mlsanity.com/paper/2403.11145/) (2024)
- [Exploiting Sentiment and Common Sense for Zero-shot Stance Detection](https://mlsanity.com/paper/2208.08797/) (2022)
- [Stance in Replies and Quotes (SRQ): A New Dataset For Learning Stance in Twitter Conversations](https://mlsanity.com/paper/2006.00691/) (2020)
- [Multi-Granular Multimodal Clue Fusion for Meme Understanding](https://mlsanity.com/paper/2503.12560/) (2025)
- [Multimodal Multi-turn Conversation Stance Detection: A Challenge Dataset and Effective Model](https://mlsanity.com/paper/2409.00597/) (2024)
- [Zero-Shot Conversational Stance Detection: Dataset and Approaches](https://mlsanity.com/paper/2506.17693/) (2025)
- [Can Large Language Models Address Open-Target Stance Detection?](https://mlsanity.com/paper/2409.00222/) (2024)
- [MultiClimate: Multimodal Stance Detection on Climate Change Videos](https://mlsanity.com/paper/2409.18346/) (2024)
- [Mitigating Biases of Large Language Models in Stance Detection with Counterfactual Augmented Calibration](https://mlsanity.com/paper/2402.14296/) (2024)
- [Is External Information Useful for Stance Detection with LLMs?](https://mlsanity.com/paper/2507.01543/) (2025)
- [MM-StanceDet: Retrieval-Augmented Multi-modal Multi-agent Stance Detection](https://mlsanity.com/paper/2604.27934/) (2026)
- [MIND Your Reasoning: A Meta-Cognitive Intuitive-Reflective Network for Dual-Reasoning in Multimodal Stance Detection](https://mlsanity.com/paper/2511.06057/) (2025)
- [Dynamic Emotion and Personality Profiling for Multimodal Deception Detection](https://mlsanity.com/paper/2604.17037/) (2026)

## Abstract (arXiv)

Dynamic stance classification models how a reply responds to its direct parent message, rather than how a post relates to a fixed topic. Existing work has mainly studied this problem in text-only settings, while social media interactions increasingly rely on images, screenshots, memes, reaction images, and cross-modal references. We introduce MMDS-Bench, a diagnostic benchmark for multimodal dynamic stance classification in social media parent-reply interactions. MMDS-Bench contains 3,482 multimodal instances annotated with a seven-label dynamic stance taxonomy, together with an 800-instance diagnostic subset that requires structured reasoning over parent understanding, reply understanding, and stance-relation inference. We further annotate each instance with five challenge factors covering multimodal fusion, parent framing, non-literal expression, interaction reasoning, and label-boundary ambiguity. We evaluate 12 closed-source and open-source multimodal large language models and propose a reference-grounded LLM-judge protocol for assessing reasoning quality. Results show that current MLLMs still struggle with multimodal dynamic stance understanding, especially in cases that require relational inference beyond separate parent and reply comprehension.