# MOBA-VL: Event-Localized Multi-Turn Reinforcement Learning for Real-Time MOBA Commentary

arXiv: https://arxiv.org/abs/2609.38428 · HTML: https://mlsanity.com/paper/2609.38428/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 29 references that are in the map.
- First posted on arXiv 29 Sep 2026.

## Placed in

- Multimodal LLMs / vision-language models

## Builds on (most cited first, 5 of 5)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence](https://mlsanity.com/paper/2606.19348/) (2026)
- [OpenHelix: A Short Survey, Empirical Analysis, and Open-Source Dual-System VLA Model for Robotic Manipulation](https://mlsanity.com/paper/2505.03912/) (2025)
- [DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression](https://mlsanity.com/paper/2609.19969/) (2026)
- [Game-MUG: Multimodal Oriented Game Situation Understanding and Commentary Generation Dataset](https://mlsanity.com/paper/2404.19175/) (2024)

## Abstract (arXiv)

Real-time commentary for Multiplayer Online Battle Arena (MOBA) esports requires a vision-language model (VLM) to narrate a live match second by second, both fluently and accurately. Existing streaming VLMs sound natural but often miss key events such as kills and objectives. To address this limitation, we use game telemetry, which records exactly when each event occurs, as a supervision signal. We introduce MOBA-VL, a 9B-parameter model trained on this signal with event-localized multi-turn reinforcement learning, which rewards the turns that describe each event. We also collect MOBACast, 860 professional matches (about 460 hours) across three MOBA games with word-level timestamped commentary, and MOBACast-Bench, a benchmark from held-out tournaments. On MOBACast-Bench, MOBA-VL achieves the highest Overall score on full matches (63.25 vs. 55.12 for StreamingVLM) and clips (63.45 vs. 56.22 for DeepSeek-V4.1-Flash). Event-localized credit also raises event recall from 34.5 to 42.1 over supervised fine-tuning. Code and data will be released, and demos are available on an anonymous project page at https://moba-vl.github.io.