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
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add T-GRPO RL stage (ablation) | MRE · CronQuestions | Hits@1: 0.921 -> 0.982 | Improved |
| add cold-start SFT (ablation) | MRE · CronQuestions | Hits@1: 0.904 -> 0.982 | Improved |
| add multi-trajectory sampling (ablation) | MRE · CronQuestions | Hits@1: 0.876 -> 0.982 | Improved |
| use multi-hop instead of single-hop reasoning (ablation) | MRE · CronQuestions | Hits@1: 0.812 -> 0.982 | Improved |
| T-GRPO instead of PPO (10k samples) (ablation) | MRE · CronQuestions (10k samples) | Hits@1: 0.827 -> 0.902 | Improved |
| T-GRPO tree reward vs flat GRPO (10k samples) (ablation) | MRE · CronQuestions (10k samples) | Hits@1: 0.845 -> 0.902 | Improved |
| T-GRPO vs flat GRPO (50k samples) (ablation) | MRE · CronQuestions (50k samples) | Hits@1: 0.932 -> 0.968 | Improved |
| MRE vs TempoQR on complex questions | MRE · CronQuestions (complex) | Hits@1: 0.864 -> 0.970 | Improved |
| MRE vs TempoQR on 3-hop questions | MRE · CronQuestions (3-hop) | Hits@1: 0.754 -> 0.943 | Improved |
| MRE vs TwiRGCN overall | MRE · TimeQuestions | Hits@1: 0.605 -> 0.594 | Mixed |
Closest papers by meaning
- Question Calibration and Multi-Hop Modeling for Temporal Question Answering (AAAI 2024 · similarity 0.90)
- GaussianPath:A Bayesian Multi-Hop Reasoning Framework for Knowledge Graph Reasoning (AAAI 2021 · similarity 0.90)
- DTKG: Dual-Track Knowledge Graph-Verified Reasoning Framework for Multi-Hop QA (ICML 2026 · similarity 0.90)
- Knowledge Editing with Dynamic Knowledge Graphs for Multi-Hop Question Answering (AAAI 2025 · similarity 0.88)
- LoongRL: Reinforcement Learning for Advanced Reasoning over Long Contexts (ICLR 2026 · similarity 0.88)
- Tree-of-Reasoning Question Decomposition for Complex Question Answering with Large Language Models (AAAI 2024 · similarity 0.88)
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.
