# Dynamic Updates for Language Adaptation in Visual-Language Tracking

arXiv: https://arxiv.org/abs/2503.06621 · HTML: https://mlsanity.com/paper/2503.06621/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Mar 2025.

## Built on by (most cited first, 18 of 18)

- [ReasoningTrack: Chain-of-Thought Reasoning for Long-term Vision-Language Tracking](https://mlsanity.com/paper/2508.05221/) (2025)
- [Bridging Time and Space: Decoupled Spatio-Temporal Alignment for Video Grounding](https://mlsanity.com/paper/2604.08014/) (2026)
- [RAGTrack: Language-aware RGBT Tracking with Retrieval-Augmented Generation](https://mlsanity.com/paper/2603.03617/) (2026)
- [4DVLT: Dynamic Scene Understanding with Worldline-Centered Vision-Language Tracking](https://mlsanity.com/paper/2606.22631/) (2026)
- [Learning Frequency and Memory-Aware Prompts for Multi-Modal Object Tracking](https://mlsanity.com/paper/2506.23972/) (2025)
- [EPIPTrack: Rethinking Prompt Modeling with Explicit and Implicit Prompts for Multi-Object Tracking](https://mlsanity.com/paper/2510.13235/) (2025)
- [How Far are Modern Trackers from UAV-Anti-UAV? A Million-Scale Benchmark and New Baseline](https://mlsanity.com/paper/2512.07385/) (2025)
- [Learning to Track Instance from Single Nature Language Description](https://mlsanity.com/paper/2605.07064/) (2026)
- [SENTRY: SAM2-Enhanced Neighbor-Aware and Temporally Reasoned Memory for Visual Tracking](https://mlsanity.com/paper/2606.24449/) (2026)
- [Semantic-Aware Temporal Adaptation for UAV Anti-UAV Tracking](https://mlsanity.com/paper/2607.26511/) (2026)
- [SOI is the Root of All Evil: Quantifying and Breaking Similar Object Interference in Single Object Tracking](https://mlsanity.com/paper/2508.09524/) (2025)
- [MUOT_3M: A 3 Million Frame Multimodal Underwater Benchmark and the MUTrack Tracking Method](https://mlsanity.com/paper/2602.18006/) (2026)
- [Interactive Tracking: A Human-in-the-Loop Paradigm with Memory-Augmented Adaptation](https://mlsanity.com/paper/2604.01974/) (2026)
- [MPerS: Dynamic MLLM MixExperts Perception-Guided Remote Sensing Scene Segmentation](https://mlsanity.com/paper/2605.10769/) (2026)
- [Dynamic Parsing and Updating Natural Language Specification using VLMs for Robust Vision-Language Tracking](https://mlsanity.com/paper/2606.29357/) (2026)
- [AnyTrack: Unifying Visual Object Tracking with Any Modalities](https://mlsanity.com/paper/2608.06773/) (2026)
- [Efficient Language-to-Vision Feature Injection for Referring Single-Object Tracking](https://mlsanity.com/paper/2608.29126/) (2026)
- [Difference Feature Map Distillation: Transferring Inter-Sample Relational Knowledge Towards Efficient Transformer-Based Tracking](https://mlsanity.com/paper/2610.05707/) (2026)

## Abstract (arXiv)

The consistency between the semantic information provided by the multi-modal reference and the tracked object is crucial for visual-language (VL) tracking. However, existing VL tracking frameworks rely on static multi-modal references to locate dynamic objects, which can lead to semantic discrepancies and reduce the robustness of the tracker. To address this issue, we propose a novel vision-language tracking framework, named DUTrack, which captures the latest state of the target by dynamically updating multi-modal references to maintain consistency. Specifically, we introduce a Dynamic Language Update Module, which leverages a large language model to generate dynamic language descriptions for the object based on visual features and object category information. Then, we design a Dynamic Template Capture Module, which captures the regions in the image that highly match the dynamic language descriptions. Furthermore, to ensure the efficiency of description generation, we design an update strategy that assesses changes in target displacement, scale, and other factors to decide on updates. Finally, the dynamic template and language descriptions that record the latest state of the target are used to update the multi-modal references, providing more accurate reference information for subsequent inference and enhancing the robustness of the tracker. DUTrack achieves new state-of-the-art performance on four mainstream vision-language and two vision-only tracking benchmarks, including LaSOT, LaSOT_rmext, TNL2K, OTB99-Lang, GOT-10K, and UAV123. Code and models are available at https://github.com/GXNU-ZhongLab/DUTrack.