Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.28828 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910267205484544 |
|---|---|
| author | Feng, Yujie Li, Jian Zhou, Zhihan Xu, Pengfei Zhang, Yujia Li, Xiaoyu Zhou, Xiaohui Zhao, Alan Chen, Xi Wu, Xiao-Ming |
| author_facet | Feng, Yujie Li, Jian Zhou, Zhihan Xu, Pengfei Zhang, Yujia Li, Xiaoyu Zhou, Xiaohui Zhao, Alan Chen, Xi Wu, Xiao-Ming |
| contents | Large Language Models (LLMs) achieve impressive performance across many tasks but remain prone to hallucination, especially in long-form generation where redundant retrieved contexts and lengthy reasoning chains amplify factual errors. Recent studies highlight a critical phenomenon: the closer key information appears to the model outputs, the higher the factual accuracy. However, existing retrieval-augmented language models (RALMs) lack effective mechanisms to ensure this proximity - external evidence is injected into reasoning via multi-turn retrieval, but this cannot ensure key information stays close to the outputs. We propose Micro-Macro Retrieval (M2R), a novel retrieve-while-generate framework to fill this gap. At the macro level, M2R retrieves coarse-grained evidence from external sources; at the micro level, it extracts essential results from a key information repository built during reasoning and reuses them while generating answers. This design directly addresses the key-information-to-output proximity bottleneck, effectively reducing hallucination in long-form tasks. M2R is trained with a curriculum learning-based reinforcement learning strategy using customized rule-based rewards, enabling stable acquisition of retrieval and grounding skills. Extensive experiments across different benchmarks demonstrate the effectiveness of M2R, especially in lengthy-context settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_28828 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Micro-Macro Retrieval: Reducing Long-Form Hallucination in Large Language Models Feng, Yujie Li, Jian Zhou, Zhihan Xu, Pengfei Zhang, Yujia Li, Xiaoyu Zhou, Xiaohui Zhao, Alan Chen, Xi Wu, Xiao-Ming Computation and Language Artificial Intelligence Large Language Models (LLMs) achieve impressive performance across many tasks but remain prone to hallucination, especially in long-form generation where redundant retrieved contexts and lengthy reasoning chains amplify factual errors. Recent studies highlight a critical phenomenon: the closer key information appears to the model outputs, the higher the factual accuracy. However, existing retrieval-augmented language models (RALMs) lack effective mechanisms to ensure this proximity - external evidence is injected into reasoning via multi-turn retrieval, but this cannot ensure key information stays close to the outputs. We propose Micro-Macro Retrieval (M2R), a novel retrieve-while-generate framework to fill this gap. At the macro level, M2R retrieves coarse-grained evidence from external sources; at the micro level, it extracts essential results from a key information repository built during reasoning and reuses them while generating answers. This design directly addresses the key-information-to-output proximity bottleneck, effectively reducing hallucination in long-form tasks. M2R is trained with a curriculum learning-based reinforcement learning strategy using customized rule-based rewards, enabling stable acquisition of retrieval and grounding skills. Extensive experiments across different benchmarks demonstrate the effectiveness of M2R, especially in lengthy-context settings. |
| title | Micro-Macro Retrieval: Reducing Long-Form Hallucination in Large Language Models |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2605.28828 |