RAG & Retrieval

LAMAR is a language-aware multilingual reranker designed for RAG pipelines where relevant documents may appear in many languages. It aims to preserve semantic relevance while favoring documents written in the same language as the query.

The Vergecast frames “Google Zero” as a turning point for publishers and websites: the long-running exchange of crawlable content for search traffic is becoming unstable in the age of AI summaries.

The paper argues that as LLMs and agents become the main consumers of search results, retrieval quality should be judged at the document-set level rather than by isolated relevance scores.

AutoIndex argues that retrieval quality can be improved before the retriever ever runs: by learning executable programs that transform documents prior to indexing. With BM25 kept fixed, the framework improves performance across all eight CRUMB retrieval tasks.

RAGU is an open-source GraphRAG engine that separates extraction from graph consolidation, aiming to make graph-based retrieval less noisy and less expensive.

SearchOS-V1 targets a common failure mode in long-horizon web-search agents: losing track of progress and repeating unproductive searches. It introduces an explicit state layer for evidence, coverage, tasks, and failures across collaborating agents.

GRASP addresses a central challenge in Agentic RAG: deciding when to retrieve, which retrieval tool to use, and how much context to read. The framework uses reinforcement learning to coordinate semantic search, keyword search, and paragraph reading during multi-step reasoning.

NVIDIA has released the Nemotron 3 Embed family on Hugging Face, positioning it for enterprise RAG, code retrieval, agent memory, and production-scale search. The 8B BF16 model ranks first on RTEB, while the 1B variants focus on efficiency and high-throughput deployment.

A new arXiv paper introduces PAT, a RAG-based system designed to push LLM translation beyond sentence-by-sentence rendering. The study finds that specifications and corpus-informed context can encourage broader reformulation, though the benefits remain inconsistent.

A new arXiv paper introduces DeepStress, a stress-testing framework for deep search agents exposed to unreliable evidence. Instead of relying on naturally occurring retrieval failures, it creates a controlled environment for probing trustworthiness, relevance, and factuality.