Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
AgriIR is a modular RAG framework that decomposes information access into query refinement, sub-query planning, retrieval, synthesis, and evaluation stages for domain-specific knowledge retrieval. This architecture enables adaptation to new knowledge verticals without architectural modifications, using smaller 1B-parameter language models. Results from a reference implementation targeting Indian agricultural information access demonstrate domain-accurate and trustworthy retrieval under constrained resources.
Domain-specific RAG doesn't need massive models: AgriIR shows how a modular architecture with smaller models can deliver accurate, trustworthy retrieval.
This paper introduces AgriIR, a configurable retrieval augmented generation (RAG) framework designed to deliver grounded, domain-specific answers while maintaining flexibility and low computational cost. Instead of relying on large, monolithic models, AgriIR decomposes the information access process into declarative modular stages -- query refinement, sub-query planning, retrieval, synthesis, and evaluation. This design allows practitioners to adapt the framework to new knowledge verticals without modifying the architecture. Our reference implementation targets Indian agricultural information access, integrating 1B-parameter language models with adaptive retrievers and domain-aware agent catalogues. The system enforces deterministic citation, integrates telemetry for transparency, and includes automated deployment assets to ensure auditable, reproducible operation. By emphasizing architectural design and modular control, AgriIR demonstrates that well-engineered pipelines can achieve domain-accurate, trustworthy retrieval even under constrained resources. We argue that this approach exemplifies ``AI for Agriculture''by promoting accessibility, sustainability, and accountability in retrieval-augmented generation systems.