DIVER: A Multi-Stage Approach for Reasoning-intensive Information Retrieval
Duolin Sun, Meixiu Long, Dan Yang, Junjie Wang, Yecheng Luo +7 more
Abstract
Retrieval-augmented generation has achieved strong performance on knowledge-intensive tasks where query-document relevance can be identified through direct lexical or semantic matches. However, many real-world queries involve abstract reasoning, analogical thinking, or multi-step inference, which existing retrievers often struggle to capture. To address this challenge, we present DIVER, a retrieval pipeline designed...
Results & Benchmarks
Benchmark data is not yet available for this paper.
Hardware Requirements
- Expect multi-day setup/compute for meaningful reproduction based on current guidance.
Best Implementation
Complex Reasoning Rag System, Agentic Rag System
- Selected AQ-MedAI/Diver as the strongest maintained implementation for new work.
- Repository activity is within the last 24 months.
Reproduction Path
- 1
Start with AQ-MedAI/Diver and validate setup instructions in README.
- 2
Reproduce the baseline result with the provided defaults before modifying hyperparameters.
- 3
Log exact dependency versions and runtime environment for reproducibility.
Additional Implementations
No additional verified repositories beyond the primary recommendation.
Hugging Face Artifacts
No direct paper-linked artifacts were found. Showing strongest curated related artifacts.