# Visual Language Tracking with Multi-modal Interaction: A Robust Benchmark

arXiv: https://arxiv.org/abs/2409.08887 · HTML: https://mlsanity.com/paper/2409.08887/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 17 references that are in the map.
- First posted on arXiv 13 Sep 2024.

## Builds on (most cited first, 12 of 12)

- [GOT-10k: A Large High-Diversity Benchmark for Generic Object Tracking in the Wild](https://mlsanity.com/paper/1810.11981/) (2018)
- [LaSOT: A High-quality Benchmark for Large-scale Single Object Tracking](https://mlsanity.com/paper/1809.07845/) (2018)
- [Towards More Flexible and Accurate Object Tracking with Natural Language: Algorithms and Benchmark](https://mlsanity.com/paper/2103.16746/) (2021)
- [Towards Unified Token Learning for Vision-Language Tracking](https://mlsanity.com/paper/2308.14103/) (2023)
- [Joint Visual Grounding and Tracking with Natural Language Specification](https://mlsanity.com/paper/2303.12027/) (2023)
- [Divert More Attention to Vision-Language Tracking](https://mlsanity.com/paper/2207.01076/) (2022)
- [Grounding-Tracking-Integration](https://mlsanity.com/paper/1912.06316/) (2019)
- [Unifying Visual and Vision-Language Tracking via Contrastive Learning](https://mlsanity.com/paper/2401.11228/) (2024)
- [DTLLM-VLT: Diverse Text Generation for Visual Language Tracking Based on LLM](https://mlsanity.com/paper/2405.12139/) (2024)
- [Global Instance Tracking: Locating Target More Like Humans](https://mlsanity.com/paper/2202.13073/) (2022)
- [BioDrone: A Bionic Drone-based Single Object Tracking Benchmark for Robust Vision](https://mlsanity.com/paper/2402.04519/) (2024)
- [SOTVerse: A User-defined Task Space of Single Object Tracking](https://mlsanity.com/paper/2204.07414/) (2022)

## Built on by (most cited first, 10 of 10)

- [NarrLV: Towards a Comprehensive Narrative-Centric Evaluation for Long Video Generation](https://mlsanity.com/paper/2507.11245/) (2025)
- [DTVLT: A Multi-modal Diverse Text Benchmark for Visual Language Tracking Based on LLM](https://mlsanity.com/paper/2410.02492/) (2024)
- [How Texts Help? A Fine-grained Evaluation to Reveal the Role of Language in Vision-Language Tracking](https://mlsanity.com/paper/2411.15600/) (2024)
- [CSTrack: Enhancing RGB-X Tracking via Compact Spatiotemporal Features](https://mlsanity.com/paper/2505.19434/) (2025)
- [ATCTrack: Aligning Target-Context Cues with Dynamic Target States for Robust Vision-Language Tracking](https://mlsanity.com/paper/2507.19875/) (2025)
- [CausalStep: A Benchmark for Explicit Stepwise Causal Reasoning in Videos](https://mlsanity.com/paper/2507.16878/) (2025)
- [Look Less, Reason More: Rollout-Guided Adaptive Pixel-Space Reasoning](https://mlsanity.com/paper/2510.01681/) (2025)
- [Select Less, Reason More: Prioritizing Evidence Purity for Video Reasoning](https://mlsanity.com/paper/2510.15440/) (2025)
- [Omni Survey for Multimodality Analysis in Visual Object Tracking](https://mlsanity.com/paper/2508.13000/) (2025)
- [VS-LLM: Visual-Semantic Depression Assessment based on LLM for Drawing Projection Test](https://mlsanity.com/paper/2508.05299/) (2025)

## Abstract (arXiv)

Visual Language Tracking (VLT) enhances tracking by mitigating the limitations of relying solely on the visual modality, utilizing high-level semantic information through language. This integration of the language enables more advanced human-machine interaction. The essence of interaction is cognitive alignment, which typically requires multiple information exchanges, especially in the sequential decision-making process of VLT. However, current VLT benchmarks do not account for multi-round interactions during tracking. They provide only an initial text and bounding box (bbox) in the first frame, with no further interaction as tracking progresses, deviating from the original motivation of the VLT task. To address these limitations, we propose a novel and robust benchmark, VLT-MI (Visual Language Tracking with Multi-modal Interaction), which introduces multi-round interaction into the VLT task for the first time. (1) We generate diverse, multi-granularity texts for multi-round, multi-modal interaction based on existing mainstream VLT benchmarks using DTLLM-VLT, leveraging the world knowledge of LLMs. (2) We propose a new VLT interaction paradigm that achieves multi-round interaction through text updates and object recovery. When multiple tracking failures occur, we provide the tracker with more aligned texts and corrected bboxes through interaction, thereby expanding the scope of VLT downstream tasks. (3) We conduct comparative experiments on both traditional VLT benchmarks and VLT-MI, evaluating and analyzing the accuracy and robustness of trackers under the interactive paradigm. This work offers new insights and paradigms for the VLT task, enabling a fine-grained evaluation of multi-modal trackers. We believe this approach can be extended to additional datasets in the future, supporting broader evaluations and comparisons of video-language model capabilities.