https://ahren09.github.io/tags/retrieval-augmented-generation/Retrieval-Augmented GenerationHugo Blox Builder (https://hugoblox.com)en-usWed, 01 Jul 2026 00:00:00 +0000https://ahren09.github.io/media/icon_hu_eee6347cbdb2cc3f.pnghttps://ahren09.github.io/tags/retrieval-augmented-generation/https://ahren09.github.io/publication/acl26_mm_bizrag/Wed, 01 Jul 2026 00:00:00 +0000https://ahren09.github.io/publication/acl26_mm_bizrag/<h2 id="abstract">Abstract</h2> <p>Enterprise question answering frequently spans heterogeneous modalities — text, tables, charts, scanned documents, and structured databases. We introduce MM-BizRAG, a multimodal retrieval-augmented generation framework designed for general purpose enterprise Q&amp;A.</p>https://ahren09.github.io/publication/acl26_sara/Wed, 01 Jul 2026 00:00:00 +0000https://ahren09.github.io/publication/acl26_sara/<h2 id="abstract">Abstract</h2> <p>Retrieval-augmented Generation (RAG) extends large language models with external knowledge, but balancing local factual precision with global knowledge coverage under strict context budgets remains a fundamental challenge. We propose SARA, a unified RAG framework that combines fine-grained natural-language spans with compact, interpretable semantic compression vectors. SARA introduces an iterative context refinement mechanism that uses compression vectors for dynamic reranking, reducing document redundancy while maximizing query informativeness. Across multiple datasets and open-source LLM families (Mistral, Llama, Gemma), SARA delivers consistent performance gains over strong RAG baselines.</p> <h2 id="links">Links</h2> <ul> <li> </li> </ul>