GPT在生成文本时，隐含学习了因果世界模型，可判断动作序列是否合理。
A Causal World Model Underlying Next Token Prediction: Exploring GPT in a Controlled Environment
- 从注意力机制推导出因果解释，构建GPT的隐式世界模型
- 在国际象棋和五子棋模拟中，对随机合法走法预测准确率高
- 可零样本检测序列合理性，适合验证生成逻辑可靠性
生成式预训练变换器（GPT）仅通过预测下一个词进行训练，是否隐式学习了一个由序列生成的世界模型？我们通过推导GPT注意力机制的因果解释，提出一个由此衍生的因果世界模型。进一步提出，在推理阶段，GPT可被用于输入序列的零样本因果结构学习，并引入相应置信度评分。在受控环境中，使用奥赛罗和国际象棋策略游戏设置进行实证测试。将一个在真实对弈数据上预训练的GPT模型，应用于分布外的合成数据——即由随机合法走法组成的序列。结果表明，当注意力机制编码了因果结构时，该模型能以高置信度生成合法下一步动作；若生成非法动作，则未能捕捉因果结构。
原文摘要 · Abstract (English)
Are generative pre-trained transformer (GPT) models, trained only to predict the next token, implicitly learning a world model from which sequences are generated one token at a time? We address this question by deriving a causal interpretation of the attention mechanism in GPT and presenting a causal world model that arises from this interpretation. Furthermore, we propose that GPT models, at inference time, can be utilized for zero-shot causal structure learning for input sequences, and introduce a corresponding confidence score. Empirical tests were conducted in controlled environments using the setups of the Othello and Chess strategy games. A GPT, pre-trained on real-world games played with the intention of winning, was tested on out-of-distribution synthetic data consisting of sequences of random legal moves. We find that the GPT model is likely to generate legal next moves for out-of-distribution sequences for which a causal structure is encoded in the attention mechanism with high confidence. In cases where it generates illegal moves, it also fails to capture a causal structure.
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。