Scaling Test-Time Inference with Policy-Optimized, Dynamic Retrieval-Augmented Generation via KV Caching and Decoding
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910954571169792 |
|---|---|
| author | Srinivas, Sakhinana Sagar Das, Akash Gupta, Shivam Runkana, Venkataramana |
| author_facet | Srinivas, Sakhinana Sagar Das, Akash Gupta, Shivam Runkana, Venkataramana |
| contents | We present a comprehensive framework for enhancing Retrieval-Augmented Generation (RAG) systems through dynamic retrieval strategies and reinforcement fine-tuning. This approach significantly improves large language models on knowledge-intensive tasks, including opendomain question answering and complex reasoning. Our framework integrates two complementary techniques: Policy-Optimized RetrievalAugmented Generation (PORAG), which optimizes the use of retrieved information, and Adaptive Token-Layer Attention Scoring (ATLAS), which dynamically determines retrieval timing and content based on contextual needs. Together, these techniques enhance both the utilization and relevance of retrieved content, improving factual accuracy and response quality. Designed as a lightweight solution compatible with any Transformer-based LLM without requiring additional training, our framework excels in knowledge-intensive tasks, boosting output accuracy in RAG settings. We further propose CRITIC, a novel method to selectively compress key-value caches by token importance, mitigating memory bottlenecks in long-context applications. The framework also incorporates test-time scaling techniques to dynamically balance reasoning depth and computational resources, alongside optimized decoding strategies for faster inference. Experiments on benchmark datasets show that our framework reduces hallucinations, strengthens domain-specific reasoning, and achieves significant efficiency and scalability gains over traditional RAG systems. This integrated approach advances the development of robust, efficient, and scalable RAG systems across diverse applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_01281 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Scaling Test-Time Inference with Policy-Optimized, Dynamic Retrieval-Augmented Generation via KV Caching and Decoding Srinivas, Sakhinana Sagar Das, Akash Gupta, Shivam Runkana, Venkataramana Machine Learning Artificial Intelligence Computation and Language Information Retrieval We present a comprehensive framework for enhancing Retrieval-Augmented Generation (RAG) systems through dynamic retrieval strategies and reinforcement fine-tuning. This approach significantly improves large language models on knowledge-intensive tasks, including opendomain question answering and complex reasoning. Our framework integrates two complementary techniques: Policy-Optimized RetrievalAugmented Generation (PORAG), which optimizes the use of retrieved information, and Adaptive Token-Layer Attention Scoring (ATLAS), which dynamically determines retrieval timing and content based on contextual needs. Together, these techniques enhance both the utilization and relevance of retrieved content, improving factual accuracy and response quality. Designed as a lightweight solution compatible with any Transformer-based LLM without requiring additional training, our framework excels in knowledge-intensive tasks, boosting output accuracy in RAG settings. We further propose CRITIC, a novel method to selectively compress key-value caches by token importance, mitigating memory bottlenecks in long-context applications. The framework also incorporates test-time scaling techniques to dynamically balance reasoning depth and computational resources, alongside optimized decoding strategies for faster inference. Experiments on benchmark datasets show that our framework reduces hallucinations, strengthens domain-specific reasoning, and achieves significant efficiency and scalability gains over traditional RAG systems. This integrated approach advances the development of robust, efficient, and scalable RAG systems across diverse applications. |
| title | Scaling Test-Time Inference with Policy-Optimized, Dynamic Retrieval-Augmented Generation via KV Caching and Decoding |
| topic | Machine Learning Artificial Intelligence Computation and Language Information Retrieval |
| url | https://arxiv.org/abs/2504.01281 |