让游戏图像模型无视游戏风格，跨游戏通用。
Game-invariant Features Through Contrastive and Domain-adversarial Learning
- 用对比学习+对抗域分类，让特征不依赖特定游戏风格。
- 在10个游戏的1万张截图上训练几轮后，特征不再按游戏分组。
- 适合做跨游戏视觉任务，如漏洞检测，几乎不用重训。
基础游戏图像编码器常过度依赖特定游戏的视觉风格，导致在新游戏中下游任务性能下降。本文提出结合对比学习与域对抗训练的方法，学习跨游戏的视觉不变特征。通过同时鼓励相似内容聚类，并利用对抗域分类器抑制游戏特有线索，所提方法生成的嵌入能有效泛化至多样游戏。在包含10,000张截图、来自10款游戏的Bingsu数据集上的实验表明，仅经过少数训练轮次，模型特征即不再按游戏聚集，验证了成功的不变性，具备显著的跨游戏迁移潜力（如漏洞检测），且只需极少微调。该能力为无需或极少重训练即可部署的新游戏视觉模型铺平道路。
原文摘要 · Abstract (English)
Foundational game-image encoders often overfit to game-specific visual styles, undermining performance on downstream tasks when applied to new games. We present a method that combines contrastive learning and domain-adversarial training to learn game-invariant visual features. By simultaneously encouraging similar content to cluster and discouraging game-specific cues via an adversarial domain classifier, our approach produces embeddings that generalize across diverse games. Experiments on the Bingsu game-image dataset (10,000 screenshots from 10 games) demonstrate that after only a few training epochs, our model's features no longer cluster by game, indicating successful invariance and potential for improved cross-game transfer (e.g., glitch detection) with minimal fine-tuning. This capability paves the way for more generalizable game vision models that require little to no retraining on new games.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。