用古印度逻辑框架训练大模型，提升其推理可信度。
Pramana: Fine-Tuning Large Language Models for Epistemic Reasoning through Navya-Nyaya
- 用2500年历史的尼雅亚逻辑框架细调大模型，构建六阶段推理流程。
- 在55个逻辑题上达到100%语义正确率，即使格式遵循仅40%。
- 适合研究可解释推理、可信AI的学者，开源全部资源供复现。
大语言模型生成流畅文本但缺乏系统性推理能力，常自信地产生无依据的断言。苹果研究人员发现，向数学问题添加无关上下文后，LLM性能下降65%，暴露出其依赖脆弱模式匹配的本质。这种认知上的空白——无法将主张建立在可追溯证据之上——严重制约了人工智能在需论证领域的可靠性。我们提出Pramana，一种通过细调大模型学习尼雅亚（Navya-Nyaya）逻辑的新方法，这是一种延续2500年的印度推理体系。不同于通用思维链提示，尼雅亚逻辑强制执行六阶段结构化推理：SAMSHAYA（怀疑分析）、PRAMANA（证据源识别）、PANCHA AVAYAVA（含普遍规则的五成分三段论）、TARKA（反事实验证）、HETVABHASA（谬误检测）和NIRNAYA（结论确认，区分知识与假设）。该融合逻辑与认识论的框架为标准推理提供了缺失的认知支撑。我们在55个尼雅亚结构化逻辑问题（约束满足、布尔可满足性、多步推导）上对Llama 3.2-3B和DeepSeek-R1-Distill-Llama-8B进行微调。第一阶段在保留测试集上实现100%语义正确率，尽管严格格式遵循率仅为40%，表明模型已内化推理内容，即使结构控制不严。消融实验显示格式提示和温度参数对性能影响显著，最优配置随阶段变化。所有模型、数据集及训练基础设施均已发布于Hugging Face，以促进对人工智能认知框架的进一步研究。
原文摘要 · Abstract (English)
Large language models produce fluent text but struggle with systematic reasoning, often hallucinating confident but unfounded claims. When Apple researchers added irrelevant context to mathematical problems, LLM performance degraded by 65% Apple Machine Learning Research, exposing brittle pattern-matching beneath apparent reasoning. This epistemic gap, the inability to ground claims in traceable evidence, limits AI reliability in domains requiring justification. We introduce Pramana, a novel approach that teaches LLMs explicit epistemological methodology by fine-tuning on Navya-Nyaya logic, a 2,500-year-old Indian reasoning framework. Unlike generic chain-of-thought prompting, Navya-Nyaya enforces structured 6-phase reasoning: SAMSHAYA (doubt analysis), PRAMANA (evidence source identification), PANCHA AVAYAVA (5-member syllogism with universal rules), TARKA (counterfactual verification), HETVABHASA (fallacy detection), and NIRNAYA (ascertainment distinguishing knowledge from hypothesis). This integration of logic and epistemology provides cognitive scaffolding absent from standard reasoning approaches. We fine-tune Llama 3.2-3B and DeepSeek-R1-Distill-Llama-8B on 55 Nyaya-structured logical problems (constraint satisfaction, Boolean SAT, multi-step deduction). Stage 1 achieves 100% semantic correctness on held-out evaluation despite only 40% strict format adherence revealing that models internalize reasoning content even when structural enforcement is imperfect. Ablation studies show format prompting and temperature critically affect performance, with optimal configurations differing by stage. We release all models, datasets, and training infrastructure on Hugging Face to enable further research on epistemic frameworks for AI reasoning.
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。