# DoublesEval: Diagnosing Multi-Agent Tactical Reasoning in Vision-Language Models via Professional Doubles Badminton

arXiv: https://arxiv.org/abs/2608.24439 · HTML: https://mlsanity.com/paper/2608.24439/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 24 of its 31 references that are in the map.
- First posted on arXiv 25 Aug 2026.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)
- Multimodal LLMs / vision-language models / [Embodied and 3D vision-language models](https://mlsanity.com/mllm/embodied-and-3d-vision-language-models/)
- LLM agents and tool use / [Multimodal and embodied agents](https://mlsanity.com/agents/multimodal-and-embodied-agents/)

## Builds on (most cited first, 24 of 24)

- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [The Kinetics Human Action Video Dataset](https://mlsanity.com/paper/1705.06950/) (2017)
- [Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis](https://mlsanity.com/paper/2405.21075/) (2024)
- [Video-LLaVA: Learning United Visual Representation by Alignment Before Projection](https://mlsanity.com/paper/2311.10122/) (2023)
- [OpenAI GPT-5 System Card](https://mlsanity.com/paper/2601.03267/) (2025)
- [Eyes Wide Shut? Exploring the Visual Shortcomings of Multimodal LLMs](https://mlsanity.com/paper/2401.06209/) (2024)
- [When and why vision-language models behave like bags-of-words, and what to do about it?](https://mlsanity.com/paper/2210.01936/) (2022)
- [Winoground: Probing Vision and Language Models for Visio-Linguistic Compositionality](https://mlsanity.com/paper/2204.03162/) (2022)
- [VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding](https://mlsanity.com/paper/2501.13106/) (2025)
- [TempCompass: Do Video LLMs Really Understand Videos?](https://mlsanity.com/paper/2403.00476/) (2024)
- [Vision language models are blind: Failing to translate detailed visual features into words](https://mlsanity.com/paper/2407.06581/) (2024)
- [Molmo2: Open Weights and Data for Vision-Language Models with Video Understanding and Grounding](https://mlsanity.com/paper/2601.10611/) (2026)
- [SoccerNet-v2: A Dataset and Benchmarks for Holistic Understanding of Broadcast Soccer Videos](https://mlsanity.com/paper/2011.13367/) (2020)
- [Do Vision-Language Models Represent Space and How? Evaluating Spatial Frame of Reference Under Ambiguities](https://mlsanity.com/paper/2410.17385/) (2024)
- [TrackNet: A Deep Learning Network for Tracking High-speed and Tiny Objects in Sports Applications](https://mlsanity.com/paper/1907.03698/) (2019)
- [MF-MOS: A Motion-Focused Model for Moving Object Segmentation](https://mlsanity.com/paper/2401.17023/) (2024)
- [Towards Structured Analysis of Broadcast Badminton Videos](https://mlsanity.com/paper/1712.08714/) (2017)
- [ShuttleSet: A Human-Annotated Stroke-Level Singles Dataset for Badminton Tactical Analysis](https://mlsanity.com/paper/2306.04948/) (2023)
- [MambaMOS: LiDAR-based 3D Moving Object Segmentation with Motion-aware State Space Model](https://mlsanity.com/paper/2404.12794/) (2024)
- [FineBadminton: A Multi-Level Dataset for Fine-Grained Badminton Video Understanding](https://mlsanity.com/paper/2508.07554/) (2025)
- [Scale, Don't Fine-tune: Guiding Multimodal LLMs for Efficient Visual Place Recognition at Test-Time](https://mlsanity.com/paper/2509.02129/) (2025)
- [Beyond First-Order: Learning Riemannian Geometries for Invariant Visual Place Recognition](https://mlsanity.com/paper/2602.00841/) (2026)

## Abstract (arXiv)

Visual Language Models (VLMs) excel at describing visible scene content but struggle to reason about dynamic multi-agent interactions, where action semantics depend on coordinated roles and spatial-temporal dependencies. We formalize this capability as multi-agent tactical reasoning and introduce DoublesEval, a diagnostic evaluation framework that leverages professional doubles badminton as a structurally tractable testbed. DoublesEval employs a key-moment-based protocol that decomposes rallies into tactically salient instants and probes models across four interpretable dimensions: atomic recognition, intra-segment composite understanding, cross-segment causal reasoning, and high-level tactical abstraction. This design isolates where reasoning fails, rather than merely measuring answer correctness. To address observed failure modes, we propose TacticCheck, a lightweight constraint-guided test-time consistency checker that reranks candidate answers using the model's own lower-level tactical predictions, requiring no parameter updates or ground-truth labels at inference time. Evaluating four representative open-source VLMs on 60 curated rallies (yielding ~9.6K structured instances) via a zero-shot protocol, we find that models remain weak across all diagnostic levels, with especially clear bottlenecks in spatial state, interaction binding, and terminal evidence. TacticCheck delivers consistent gains across all evaluated models, while still leaving a substantial gap to robust tactical reasoning. These results highlight the need for structured, interaction-aware evaluation paradigms for next-generation VLMs. The source code is available in hrefhttps://github.com/Chengjt1999/DoublesEvaltextcolorblueour GitHub repository.