AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Reinforcement Learning Enhanced Multi-hop Reasoning for Temporal Knowledge Question Answering

Wuzhenghong Wen, Chao Xue, Su Pan, Yuwei Sun, Minlong Peng

AAAI 2026 · Large language models › Retrieval and knowledge · approach: Knowledge graph reasoning with LLMs

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add T-GRPO RL stage (ablation)MRE · CronQuestionsHits@1: 0.921 -> 0.982Improved
add cold-start SFT (ablation)MRE · CronQuestionsHits@1: 0.904 -> 0.982Improved
add multi-trajectory sampling (ablation)MRE · CronQuestionsHits@1: 0.876 -> 0.982Improved
use multi-hop instead of single-hop reasoning (ablation)MRE · CronQuestionsHits@1: 0.812 -> 0.982Improved
T-GRPO instead of PPO (10k samples) (ablation)MRE · CronQuestions (10k samples)Hits@1: 0.827 -> 0.902Improved
T-GRPO tree reward vs flat GRPO (10k samples) (ablation)MRE · CronQuestions (10k samples)Hits@1: 0.845 -> 0.902Improved
T-GRPO vs flat GRPO (50k samples) (ablation)MRE · CronQuestions (50k samples)Hits@1: 0.932 -> 0.968Improved
MRE vs TempoQR on complex questionsMRE · CronQuestions (complex)Hits@1: 0.864 -> 0.970Improved
MRE vs TempoQR on 3-hop questionsMRE · CronQuestions (3-hop)Hits@1: 0.754 -> 0.943Improved
MRE vs TwiRGCN overallMRE · TimeQuestionsHits@1: 0.605 -> 0.594Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy