迈向通用因果推理器
Towards a Universal Causal Reasoner
AI总结：
提出UniCo数据生成框架，覆盖Pearl因果阶梯的18种查询类型，将符号示例转化为代码和自然语言，通过监督微调显著提升LLM的因果推理能力和推理忠实度。
摘要：
尽管因果推理的重要性不言而喻，但训练LLM进行因果推理仍未被充分探索。现有的数据工作大多集中在针对因果关系的特定方面对LLM进行基准测试，这使得它们不太适合训练可泛化的因果推理器。为了解决这个问题，我们提出了UniCo，一个数据生成框架，它既(1)涵盖了Pearl因果阶梯中的18种因果查询类型，又(2)将原生符号示例转化为代码和自然语言形式，以模拟因果术语未明确指定的真实世界用例。为确保数据质量，UniCo用精确的因果推理来支撑答案，并过滤掉存在推理捷径的案例。通过使用66.6K个UniCo生成的实例进行监督微调，Qwen3-4B、Qwen3-8B和Olmo-3-7B-Instruct在所有18种分布内查询类型上平均提升了22.9%，在训练分布之外的7个已建立的因果基准上，相比最先进的因果数据生成框架提升了8.1%。更重要的是，在真实世界的医学理解、法律决策和表格推理中，UniCo训练的模型始终展现出更忠实的推理轨迹，在忠实度指标上平均超过基础模型20.2%。这些结果表明，以因果为中心的训练不仅增强了因果推理能力，还赋予了LLM在一般推理任务中的因果思维。
英文摘要：
Despite the importance of causal reasoning, training LLMs to reason causally remains underexplored. Existing data efforts mostly focus on benchmarking LLMs on specific aspects of causality, making them less suitable for training generalizable causal reasoners. To address this, we propose UniCo, a data generation framework that both (1) addresses 18 causal query types across Pearl's Causal Ladder and (2) translates natively symbolic examples into code and natural language forms to simulate real-world use cases where causal terms are not explicitly specified. To ensure data quality, UniCo grounds answers with exact causal inference and filters cases with reasoning shortcuts. Upon supervised finetuning with 66.6K UniCo-generated instances, Qwen3-4B, Qwen3-8B and Olmo-3-7B-Instruct achieve an average of 22.9% improvements across all 18 in-distribution query types, and 8.1% over state-of-the-art causal data generation frameworks on 7 established causal benchmarks outside the training distribution. More importantly, in real-world medical understanding, legal decision, and tabular reasoning, UniCo-trained models consistently display more faithful reasoning traces, outperforming the base models by an average of 20.2% in faithfulness metrics. These suggest that causality-centered training not only strengthens causal reasoning, but also equips LLMs with a causal mindset in general reasoning tasks.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
