ALTo: Adaptive-Length Tokenizer for Autoregressive Mask Generation

Lingfeng Wang, Hualing Lin, Senda Chen, Tao Wang, Changxu Cheng, Yangyang Zhong, Dong Zheng, Wuyue Zhao

Keywords: multimodal generation, multimodal large language model, adaptive-length

TL;DR: ALTo is a novel framework leveraging adaptive-length tokens for multimodal mask generation, based on MLLM

Abstract: While humans effortlessly draw visual objects and shapes by adaptively allocating attention based on their complexity, existing multimodal large language models (MLLMs) remain constrained by rigid token representations. Bridging this gap, we propose ALTo, an adaptive length tokenizer for autoregressive mask generation. To achieve this, a novel token length predictor is designed, along with a length regularization term and a differentiable token chunking strategy. We further build ALToLLM that seamlessly integrates ALTo into MLLM. Preferences on the trade-offs between mask quality and efficiency is implemented by group relative policy optimization (GRPO). Experiments demonstrate that ALToLLM achieves state-of-the-art performance with adaptive token cost on popular segmentation benchmarks. Code and models will be released.

Primary Area: Deep learning (e.g., architectures, generative models, optimization for deep networks, foundation models, LLMs)

Submission Number: 3412