论文标题
保持冷静和探索：基于文本的游戏中动作的语言模型
Keep CALM and Explore: Language Models for Action Generation in Text-based Games
论文作者
论文摘要
基于文本的游戏为自主代理人以自然语言运作并处理巨大的动作空间提出了一个独特的挑战。在本文中，我们提出上下文动作语言模型（平静），以在每个游戏状态下生成一套紧凑的动作候选者。我们的主要见解是在人类游戏玩法上训练语言模型，在该游戏中，人们展示了语言先验和一般的游戏意识，以有希望的动作以游戏历史为条件。我们将平静与加强学习代理人相结合，将生成的动作候选者重新排列以最大程度地提高游戏中的回报。我们使用耶利哥基准测试，在训练期间没有平静的游戏中评估我们的方法。与以前的最新模型相比，我们的方法获得了平均游戏得分的69％相对提高。令人惊讶的是，在其中一半的游戏中，平静比其他模型具有竞争地面真理动作的竞争能力或更好。代码和数据可在https://github.com/princeton-nlp/calm-textgame上找到。
Text-based games present a unique challenge for autonomous agents to operate in natural language and handle enormous action spaces. In this paper, we propose the Contextual Action Language Model (CALM) to generate a compact set of action candidates at each game state. Our key insight is to train language models on human gameplay, where people demonstrate linguistic priors and a general game sense for promising actions conditioned on game history. We combine CALM with a reinforcement learning agent which re-ranks the generated action candidates to maximize in-game rewards. We evaluate our approach using the Jericho benchmark, on games unseen by CALM during training. Our method obtains a 69% relative improvement in average game score over the previous state-of-the-art model. Surprisingly, on half of these games, CALM is competitive with or better than other models that have access to ground truth admissible actions. Code and data are available at https://github.com/princeton-nlp/calm-textgame.
