GAM-RAG: Gain-Adaptive Memory for Evolving Retrieval in Retrieval-Augmented Generation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914363225407488 |
|---|---|
| author | Wang, Yifan Jiang, Mingxuan Sun, Zhihao Cao, Yixin Liu, Yicun Chen, Keyang Ye, Guangnan Chai, Hongfeng |
| author_facet | Wang, Yifan Jiang, Mingxuan Sun, Zhihao Cao, Yixin Liu, Yicun Chen, Keyang Ye, Guangnan Chai, Hongfeng |
| contents | Retrieval-Augmented Generation (RAG) grounds large language models with external evidence, but many implementations rely on pre-built indices that remain static after construction. Related queries therefore repeat similar multi-hop traversal, increasing latency and compute. Motivated by schema-based learning in cognitive neuroscience, we propose GAM-RAG, a training-free framework that accumulates retrieval experience from recurring or related queries and updates retrieval memory over time. GAM-RAG builds a lightweight, relation-free hierarchical index whose links capture potential co-occurrence rather than fixed semantic relations. During inference, successful retrieval episodes provide sentence-level feedback, updating sentence memories so evidence useful for similar reasoning types becomes easier to activate later. To balance stability and adaptability under noisy feedback, we introduce an uncertainty-aware, Kalman-inspired gain rule that jointly updates memory states and perplexity-based uncertainty estimates. It applies fast updates for reliable novel signals and conservative refinement for stable or noisy memories. We provide a theoretical analysis of the update dynamics, and empirically show that GAM-RAG improves average performance by 3.95% over the strongest baseline and by 8.19% with 5-turn memory, while reducing inference cost by 61%. Our code and datasets are available at: https://anonymous.4open.science/r/GAM_RAG-2EF6. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_01783 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | GAM-RAG: Gain-Adaptive Memory for Evolving Retrieval in Retrieval-Augmented Generation Wang, Yifan Jiang, Mingxuan Sun, Zhihao Cao, Yixin Liu, Yicun Chen, Keyang Ye, Guangnan Chai, Hongfeng Artificial Intelligence Retrieval-Augmented Generation (RAG) grounds large language models with external evidence, but many implementations rely on pre-built indices that remain static after construction. Related queries therefore repeat similar multi-hop traversal, increasing latency and compute. Motivated by schema-based learning in cognitive neuroscience, we propose GAM-RAG, a training-free framework that accumulates retrieval experience from recurring or related queries and updates retrieval memory over time. GAM-RAG builds a lightweight, relation-free hierarchical index whose links capture potential co-occurrence rather than fixed semantic relations. During inference, successful retrieval episodes provide sentence-level feedback, updating sentence memories so evidence useful for similar reasoning types becomes easier to activate later. To balance stability and adaptability under noisy feedback, we introduce an uncertainty-aware, Kalman-inspired gain rule that jointly updates memory states and perplexity-based uncertainty estimates. It applies fast updates for reliable novel signals and conservative refinement for stable or noisy memories. We provide a theoretical analysis of the update dynamics, and empirically show that GAM-RAG improves average performance by 3.95% over the strongest baseline and by 8.19% with 5-turn memory, while reducing inference cost by 61%. Our code and datasets are available at: https://anonymous.4open.science/r/GAM_RAG-2EF6. |
| title | GAM-RAG: Gain-Adaptive Memory for Evolving Retrieval in Retrieval-Augmented Generation |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2603.01783 |