A Dynamic Retrieval-augmented Generation System With Selective Memory And Remembrance

Abstract

We introduce *Adaptive RAG Memory* (ARM), a retrieval-augmented generation (RAG) framework that replaces a static vector index with a *dynamic* memory substrate governed by selective remembrance and decay. Frequently retrieved items are consolidated and protected from forgetting, while rarely used items gradually decay, inspired by cognitive consolidation and forgetting principles. On a lightweight retrieval benchmark, ARM reaches near state-of-the-art performance (e.g., NDCG@5 \(\approx\) 0.940, Recall@5 \(=1.000\)) with only \(\sim\)22M parameters in the embedding layer, achieving the best efficiency among ultra-efficient models (\(<\)25M parameters). In addition, we compare static vs. dynamic RAG combinations across Llama 3.1 and GPT-4o. Llama 3.1 with static RAG achieves the highest key-term coverage (67.2%) at moderate latency, while GPT-4o with a dynamic selective retrieval policy attains the fastest responses (8.2s on average) with competitive coverage (58.7%). We further presen

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keybursa2026a

Related papers

• RAG Without Forgetting: Continual Query-infused Key Memory (2026)0.00
• Hetarag: Hybrid Deep Retrieval-augmented Generation Across Heterogeneous Data Stores (2025)3.27
• Ragdb: A Zero-dependency, Embeddable Architecture For Multimodal Retrieval-augmented Generation On The Edge (2025)0.00
• Retro-li: Small-scale Retrieval Augmented Generation Supporting Noisy Similarity Searches And Domain Shift Generalization (2024)4.06
• Erarag: Efficient And Incremental Retrieval Augmented Generation For Growing Corpora (2025)4.51
• Optimizing Retrieval For RAG Via Reinforcement Learning (2025)0.00
• Mememo: On-device Retrieval Augmentation For Private And Personalized Text Generation (2024)10.78
• Funnelrag: A Coarse-to-fine Progressive Retrieval Paradigm For RAG (2024)3.58