MINER: Mining the Underlying Pattern of Modality-Specific Neurons in Multimodal Large Language Models

Kaichen Huang, Jiahao Huo, Yibo Yan, Kun Wang, Yutao Yue, Xuming Hu

Keywords: MLLMs, neuron analysis, interpretability

Abstract: In recent years, multimodal large language models (MLLMs) have significantly advanced, integrating more modalities into diverse applications. However, the lack of explainability remains a major barrier to their use in scenarios requiring decision transparency. Current neuron-level explanation paradigms mainly focus on knowledge localization or language- and domain-specific analyses, leaving the exploration of multimodality largely unaddressed. To tackle these challenges, we propose MINER, a transferable framework for mining modality-specific neurons (MSNs) in MLLMs, which comprises four stages: (1) modality separation, (2) importance score calculation, (3) importance score aggregation, (4) modality-specific neuron selection. Extensive experiments across six benchmarks and two MLLMs show that (1) deactivating ONLY 2% of MSNs significantly reduce MLLMs performance (0.56 to 0.24 for Qwen2-VL, 0.69 to 0.31 for Qwen2-Audio), (2) different modalities mainly converge in the lower layers, (3) MSNs influence how key information from various modalities converges to the last token, (4) We observed two intriguing phenomena, semantic probing and semantic telomeres.

Primary Area: foundation or frontier models, including LLMs

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 2490