Research map / Large language models / Retrieval and knowledge
GPS: Directed Acyclic Graph guided Proactive Information Seeking in Large Language Models
Ruiqing Li, Yifeng Xu, Xinke Jiang, Zhibang Yang, Xinyu Ma, Yue Fang, Junfeng Zhao, Yasha Wang, Xu Chu
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 clarification-oriented reinforcement learning (ablation) | GPS (Qwen2.5-7B-Instruct) · Synthetic | Success Rate (%): 52.2 -> 60.2 | Improved |
| add clarification-oriented reinforcement learning (ablation) | GPS (Qwen2.5-7B-Instruct) · Synthetic | Weighted Clarification Turns: 5.57 -> 4.59 | Improved |
| add clarification-oriented reinforcement learning (ablation) | GPS (Qwen2.5-7B-Instruct) · ConditionalQA | Success Rate (%): 67.7 -> 73.4 | Improved |
| add clarification-oriented reinforcement learning (ablation) | GPS (Qwen2.5-7B-Instruct) · ConditionalQA | Weighted Clarification Turns: 3.63 -> 2.91 | Improved |
| add efficiency reward (ablation) | GPS (Qwen2.5-7B-Instruct) · Synthetic | Success Rate (%): 59.0 -> 60.2 | Improved |
| add efficiency reward (ablation) | GPS (Qwen2.5-7B-Instruct) · Synthetic | Weighted Clarification Turns: 5.06 -> 4.59 | Improved |
| add efficiency reward (ablation) | GPS (Qwen2.5-7B-Instruct) · ConditionalQA | Success Rate (%): 70.7 -> 73.4 | Improved |
| add structural quality reward (ablation) | GPS (Qwen2.5-7B-Instruct) · Synthetic | Success Rate (%): 56.1 -> 60.2 | Improved |
| add structural quality reward (ablation) | GPS (Qwen2.5-7B-Instruct) · Synthetic | Weighted Clarification Turns: 5.32 -> 4.59 | Improved |
| add structural quality reward (ablation) | GPS (Qwen2.5-7B-Instruct) · ConditionalQA | Success Rate (%): 70.3 -> 73.4 | Improved |
| add dynamic traversal (ablation) | GPS (Qwen2.5-7B-Instruct) · Synthetic | Weighted Clarification Turns: 5.19 -> 4.59 | Improved |
| add dynamic traversal (ablation) | GPS (Qwen2.5-7B-Instruct) · ConditionalQA | Weighted Clarification Turns: 3.63 -> 2.91 | Improved |
Closest papers by meaning
- Explore-on-Graph: Incentivizing Autonomous Exploration of Large Language Models on Knowledge Graphs with Path-refined Reward Modeling (ICLR 2026 · similarity 0.89)
- Scent of Knowledge: Optimizing Search-Enhanced Reasoning with Information Foraging (NeurIPS 2025 · similarity 0.88)
- Simple is Effective: The Roles of Graphs and Large Language Models in Knowledge-Graph-Based Retrieval-Augmented Generation (ICLR 2025 · similarity 0.88)
- Adapting to Non-Stationary Environments: Multi-Armed Bandit Enhanced Retrieval-Augmented Generation on Knowledge Graphs (AAAI 2025 · similarity 0.87)
- ReSearch: Learning to Reason with Search for LLMs via Reinforcement Learning (NeurIPS 2025 · similarity 0.87)
- Think-on-Graph 2.0: Deep and Faithful Large Language Model Reasoning with Knowledge-guided Retrieval Augmented Generation (ICLR 2025 · similarity 0.87)
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.
