POINTS-Long: Adaptive Dual-Mode Visual Reasoning in MLLMs

TL;DR

POINTS-Long introduces an adaptive dual-mode multimodal model that dynamically balances efficiency and accuracy in visual reasoning, especially for long videos and streaming scenarios.

Contribution

It presents a novel dual-mode architecture with dynamic token scaling and streaming visual understanding capabilities for scalable MLLMs.

Findings

Standby mode retains 97.7-99.7% accuracy with only 1/40-1/10th tokens.

Focus mode achieves optimal performance on fine-grained visual tasks.

Supports efficient long-form visual understanding via detachable KV-cache.

Abstract

Multimodal Large Language Models (MLLMs) have recently demonstrated remarkable capabilities in cross-modal understanding and generation. However, the rapid growth of visual token sequences--especially in long-video and streaming scenarios--poses a major challenge to their scalability and real-world deployment. Thus, we introduce POINTS-Long, a native dual-mode MLLM featuring dynamic visual token scaling inspired by the human visual system. The model supports two complementary perception modes: focus mode and standby mode, enabling users to dynamically trade off efficiency and accuracy during inference. On fine-grained visual tasks, the focus mode retains the optimal performance, while on long-form general visual understanding, the standby mode retains 97.7-99.7% of the original accuracy using only 1/40-1/10th of the visual tokens. Moreover, POINTS-Long natively supports streaming visual understanding via a dynamically detachable KV-cache design, allowing efficient maintenance of ultra-long visual memory. Our work provides new insights into the design of future MLLMs and lays the foundation for adaptive and efficient long-form visual understanding.