用判别式方法让大模型更懂关系推理，避免胡编乱造。
RADAR: Reasoning as Discrimination with Aligned Representations for LLM-based Knowledge Graph Reasoning
- 把知识图谱推理转为实体选择任务，用强化学习提升实体区分度。
- 在4个基准上链接预测提升5-6%，中间表示的语义信息量提高62.9%。
- 适合需要可靠推理、避免幻觉的AI系统开发者使用。
知识图谱推理（KGR）旨在推断缺失事实，近年越来越多依赖大语言模型（LLM）的语义先验与推理能力。然而，现有生成式方法容易记忆表面共现关系，而非真正理解关系语义，限制了分布外泛化能力。为此，我们提出RADAR，将KGR从生成式模式重构为判别式关系推理。将KGR重新建模为判别性实体选择任务，通过强化学习强制提升实体间的相对可分性，而非仅模仿生成概率。利用这种可分性，推理直接在表示空间进行，确保与判别优化一致，并避开生成带来的幻觉。在四个基准测试中，RADAR相比强基线在链接预测和三元组分类上取得5-6%的相对提升，同时使中间表示中的任务相关互信息提升62.9%，表明其具备更强的鲁棒性与迁移能力。
原文摘要 · Abstract (English)
Knowledge graph reasoning (KGR) infers missing facts, with recent advances increasingly harnessing the semantic priors and reasoning abilities of Large Language Models (LLMs). However, prevailing generative paradigms are prone to memorizing surface-level co-occurrences rather than learning genuine relational semantics, limiting out-of-distribution generalization. To address this, we propose RADAR, which reformulates KGR from generative pattern matching to discriminative relational reasoning. We recast KGR as discriminative entity selection, where reinforcement learning enforces relative entity separability beyond token-likelihood imitation. Leveraging this separability, inference operates directly in representation space, ensuring consistency with the discriminative optimization and bypassing generation-induced hallucinations. Across four benchmarks, RADAR achieves 5-6% relative gains on link prediction and triple classification over strong LLM baselines, while increasing task-relevant mutual information in intermediate representations by 62.9%, indicating more robust and transferable relational 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。