Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Token Activation Map to Visually Explain Multimodal LLMs

About

Multimodal large language models (MLLMs) are broadly empowering various fields. Despite their advancements, the explainability of MLLMs remains less explored, hindering deeper understanding, model credibility, and effective visualization. Unlike conventional vision models (e.g., CNNs, ViTs, CLIP) that produce a single output, MLLMs generate sequences of tokens progressively, where each generated token depends on the previous context. Therefore, earlier context tokens can introduce redundant activations that interfere with the explanation of later tokens beyond their original information. Existing studies often overlook this issue, but our observations reveal that these redundant correlations can significantly hurt the reliability of explanations. To address this, we propose an estimated causal inference method to mitigate the interference of context to achieve high-quality MLLM explanation, with a novel rank Gaussian filter to further reduce activation noises. We term this method Token Activation Map (TAM) to highlight the consideration of interactions between tokens. TAM also indicates that it excels at explaining multiple tokens of MLLM, which is different from the Class Activation Map (CAM) for a single prediction. Our TAM method significantly outperforms existing SoTA methods, showcasing high-quality visualization results that can be utilized for various scenarios, such as object localization, failure case analysis, video visualization, MLLMs visual comparison, and model understanding (e.g., color, shape, action, location, visual reasoning, multi-turn conversation, etc). The code is available atgithub.com/xmed-lab/TAM.

Yi Li, Hualiang Wang, Xinpeng Ding, Haonan Wang, Xiaomeng Li• 2025

Related benchmarks

TaskDatasetResultRank
Referring SegmentationRefCOCO (val)--
84
Image SegmentationCOCO
mIoU25
39
Visual AttributionThinking-Model Attribution Dataset Code
LDS72
24
Visual AttributionThinking-Model Attribution Dataset General
LDS0.71
24
Visual AttributionThinking-Model Attribution Dataset Math
LDS75
24
Visual AttributionThinking-Model Attribution Dataset Science
LDS69
24
Visual AttributionThinking-Model Attribution Dataset Document
LDS0.68
24
Visual AttributionThinking-Model Attribution Dataset Overall
Avg. LDS69
24
Image Captioning ExplainabilityMS-COCO
Insertion Score0.6317
16
Hallucination InterpretationRePOPE
Insertion41.68
16
Showing 10 of 13 rows

Other info

Follow for update