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

Patch-as-Decodable-Token: Towards Unified Multi-Modal Vision Tasks in MLLMs

About

Multimodal large language models (MLLMs) have advanced rapidly in recent years. However, existing approaches for vision tasks often rely on indirect representations, such as generating coordinates as text for detection, which limits performance and prevents dense prediction tasks like segmentation. To overcome these challenges, we introduce Patch-as-Decodable Token (PaDT), a unified paradigm that enables MLLMs to directly generate both textual and diverse visual outputs. Central to PaDT are Visual Reference Tokens (VRTs), derived from visual patch embeddings of query images and interleaved seamlessly with LLM's output textual tokens. A lightweight decoder then transforms LLM's outputs into detection, segmentation, and grounding predictions. Unlike prior methods, PaDT processes VRTs independently at each forward pass and dynamically expands the embedding table, thus improving localization and differentiation among similar objects. We further tailor a training strategy for PaDT by randomly selecting VRTs for supervised fine-tuning and introducing a robust per-token cross-entropy loss. Our empirical studies across four visual perception and understanding tasks suggest PaDT consistently achieving state-of-the-art performance, even compared with significantly larger MLLM models. The code is available at https://github.com/Gorilla-Lab-SCUT/PaDT.

Yongyi Su, Haojie Zhang, Shijie Li, Nanqing Liu, Jingyi Liao, Junyi Pan, Yuan Liu, Xiaofen Xing, Chong Sun, Chen Li, Nancy F. Chen, Shuicheng Yan, Xulei Yang, Xun Xu• 2025

Related benchmarks

TaskDatasetResultRank
Referring Expression SegmentationRefCOCO (testA)
cIoU81.5
257
Referring Expression SegmentationRefCOCO+ (testA)
cIoU79.4
230
Referring Expression SegmentationRefCOCO+ (val)
cIoU77.6
223
Referring Expression SegmentationRefCOCO (testB)
cIoU82.2
213
Referring Expression SegmentationRefCOCO (val)
cIoU81.3
212
Referring Expression SegmentationRefCOCO+ (testB)
cIoU76.3
210
Referring Expression SegmentationRefCOCOg (val)
cIoU78.1
129
Referring Expression SegmentationRefCOCOg (test)
cIoU78.5
118
Anomaly SegmentationMVTec AD--
105
Pixel-level Anomaly DetectionColonDB--
39
Showing 10 of 15 rows

Other info

GitHub

Follow for update