Paper
RADAR: Reasoning as Discrimination with Aligned Representations for LLM-based Knowledge Graph Reasoning
Authors
Bo Xue, Yuan Jin, Luoyi Fu, Jiaxin Ding, Xinbing Wang
Abstract
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.
Metadata
Related papers
Vibe Coding XR: Accelerating AI + XR Prototyping with XR Blocks and Gemini
Ruofei Du, Benjamin Hersh, David Li, Nels Numan, Xun Qian, Yanhe Chen, Zhongy... • 2026-03-25
Comparing Developer and LLM Biases in Code Evaluation
Aditya Mittal, Ryan Shar, Zichu Wu, Shyam Agarwal, Tongshuang Wu, Chris Donah... • 2026-03-25
The Stochastic Gap: A Markovian Framework for Pre-Deployment Reliability and Oversight-Cost Auditing in Agentic Artificial Intelligence
Biplab Pal, Santanu Bhattacharya • 2026-03-25
Retrieval Improvements Do Not Guarantee Better Answers: A Study of RAG for AI Policy QA
Saahil Mathur, Ryan David Rittner, Vedant Ajit Thakur, Daniel Stuart Schiff, ... • 2026-03-25
MARCH: Multi-Agent Reinforced Self-Check for LLM Hallucination
Zhuo Li, Yupeng Zhang, Pengyu Cheng, Jiajun Song, Mengyu Zhou, Hao Li, Shujie... • 2026-03-25
Raw Data (Debug)
{
"raw_xml": "<entry>\n <id>http://arxiv.org/abs/2602.21951v1</id>\n <title>RADAR: Reasoning as Discrimination with Aligned Representations for LLM-based Knowledge Graph Reasoning</title>\n <updated>2026-02-25T14:34:02Z</updated>\n <link href='https://arxiv.org/abs/2602.21951v1' rel='alternate' type='text/html'/>\n <link href='https://arxiv.org/pdf/2602.21951v1' rel='related' title='pdf' type='application/pdf'/>\n <summary>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.</summary>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n <published>2026-02-25T14:34:02Z</published>\n <arxiv:primary_category term='cs.CL'/>\n <author>\n <name>Bo Xue</name>\n </author>\n <author>\n <name>Yuan Jin</name>\n </author>\n <author>\n <name>Luoyi Fu</name>\n </author>\n <author>\n <name>Jiaxin Ding</name>\n </author>\n <author>\n <name>Xinbing Wang</name>\n </author>\n </entry>"
}