用因果推理让大模型更准更快，还省token。
Debiasing Large Language Models via Adaptive Causal Prompting with Sketch-of-Thought
- 基于因果模型动态选择推理干预策略
- 比CoT少用70%以上token，准确率更高
- 适合需要高效可靠推理的场景
尽管链式思维（Chain-of-Thought, CoT）等提示方法在大语言模型（LLM）中取得进展，但现有方法仍存在令牌消耗过大、跨任务泛化能力弱的问题。为此，我们提出自适应因果提示与思维草图框架（Adaptive Causal Prompting with Sketch-of-Thought, ACPS），利用结构因果模型推断查询对答案的因果效应，并自适应选择合适的干预方式（即标准前门调整和条件前门调整）。该设计实现了无需任务特定微调的跨异构任务泛化因果推理。通过用简洁的思维草图替代冗长的CoT，ACPS显著降低令牌使用量与推理成本。在多个推理基准和大模型上的实验表明，ACPS在准确性、鲁棒性和计算效率方面均持续优于现有提示基线。
原文摘要 · Abstract (English)
Despite notable advancements in prompting methods for Large Language Models (LLMs), such as Chain-of-Thought (CoT), existing strategies still suffer from excessive token usage and limited generalisability across diverse reasoning tasks. To address these limitations, we propose an Adaptive Causal Prompting with Sketch-of-Thought (ACPS) framework, which leverages structural causal models to infer the causal effect of a query on its answer and adaptively select an appropriate intervention (i.e., standard front-door and conditional front-door adjustments). This design enables generalisable causal reasoning across heterogeneous tasks without task-specific retraining. By replacing verbose CoT with concise Sketch-of-Thought, ACPS enables efficient reasoning that significantly reduces token usage and inference cost. Extensive experiments on multiple reasoning benchmarks and LLMs demonstrate that ACPS consistently outperforms existing prompting baselines in terms of accuracy, robustness, and computational efficiency.
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。