EMNLP Findings2025
MARIO-0.5B: A Multi-Agent Lightweight Model for Real-Time Open Information Extraction in Low-Resource Settings
Donghai Zhang, Shuangtao Yang, Xiaozheng Dong, Wei Song, Bo Fu
Abstract
Large language models (LLMs) have shown remarkable capabilities in open information extraction. However, their substantial resource requirements often restrict their deployment in resource-constrained industrial settings, particularly on edge devices. The high computational demands also lead to increased latency, making them difficult to apply in real-time applications. In this paper, we introduce MARIO-0.5B, an ultra-lightweight model trained on instructionbased samples in Chinese, English, Korean, and Russian. We also present a novel multiagent framework, SMOIE, which integrates schema mining, information extraction, reasoning, and decision-making to effectively support MARIO-0.5B. The experimental results show that our framework outperforms largescale models with up to 70B parameters, reducing computational resources by 140x and delivering 11x faster response times. Moreover, it operates efficiently in CPU-only environments, which makes it well-suited for widespread industrial deployment.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- Revisiting Relation Extraction in the era of Large Language Models cites Two are Better than One: Joint Entity and Relation Extraction with Table-Sequence Encoders
- A Partition Filter Network for Joint Entity and Relation Extraction cites Two are Better than One: Joint Entity and Relation Extraction with Table-Sequence Encoders
Builds on9
- LoRA: Low-Rank Adaptation of Large Language ModelsEdward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu et al.ICLR 2022 · 18,833 citations
- Two are Better than One: Joint Entity and Relation Extraction with Table-Sequence EncodersJue Wang, Wei LuEMNLP 2020 · 209 citations
- Revisiting Relation Extraction in the era of Large Language ModelsSomin Wadhwa, Silvio Amir, Byron C. WallaceACL 2023 · 145 citations
- A Partition Filter Network for Joint Entity and Relation ExtractionZhiheng Yan, Chong Zhang, Jinlan Fu, Qi Zhang et al.EMNLP 2021 · 142 citations
- IOPO: Empowering LLMs with Complex Instruction Following via Input-Output Preference OptimizationXinghua Zhang, Haiyang Yu, Cheng Fu, Fei Huang et al.ACL 2025 · 29 citations
Related papers
- ADELIE: Aligning Large Language Models on Information ExtractionYunjia Qi, Hao Peng, Xiaozhi Wang, Bin Xu et al.EMNLP 2024 · 8 citations
- Capability Decomposition for Unified Information Extraction via Hierarchical Mixture-of-ExpertsJing Zhou, Peng Wang, Wenjun Ke, Jiajun Liu et al.ACL 2026
- Towards Scalable Lightweight GUI Agents via Multi-role OrchestrationZiwei Wang, Junjie Zheng, Leyang Yang, Sheng Zhou et al.ACL Findings 2026
- Small Models, Big Results: Achieving Superior Intent Extraction through DecompositionDanielle Cohen, Yoni Halpern, Noam Kahlon, Joel Oren et al.EMNLP 2025
- LARM: Large Auto-Regressive Model for Long-Horizon Embodied IntelligenceZhuoling Li, Xiaogang Xu, Zhenhua Xu, Ser-Nam Lim et al.ICML 2025
