analiza aiqbench
This research introduces CRINN, a framework that leverages reinforcement learning and LLMs to automatically optimize approximate nearest-neighbor search algorithms specifically for AI workloads like RAG. By treating execution speed as the reward signal, the method successfully automates complex algorithmic refinements previously requiring manual expertise, demonstrating significant performance gains on standard benchmarks.
CRINN: Contrastive Reinforcement Learning for Approximate Nearest Neighbor Search
arXiv:2508.02091v4 Announce Type: replace-cross Abstract: Approximate nearest-neighbor search (ANNS) algorithms have become increasingly critical for recent AI applications, particularly in retrieval-augmented generation (RAG) and agent-based LLM applications. In this paper, we present CRINN, a new paradigm for ANNS algorithms. CRINN treats ANNS optimization as a reinforcement learning problem where execution speed serves as the reward signal. This approach enables the automatic generation of progressively faster ANNS implementations while maintaining accuracy constraints. Our experimental evaluation demonstrates CRINN's effectiveness across six widely-used NNS benchmark datasets. When compared against state-of-the-art open-source ANNS algorithms, CRINN achieves best performance on three of them (GIST-960-Euclidean, MNIST-784-Euclidean, and GloVe-25-angular), and tied for first place on two of them (SIFT-128-Euclidean and GloVe-25-angular). The implications of CRINN's success reach well beyond ANNS optimization: It validates that LLMs augmented with reinforcement learning can function as an effective tool for automating sophisticated algorithmic optimizations that demand specialized knowledge and labor-intensive manual refinement. Code can be found at https://github.com/ornith-ai/CRINN
Abstract:Approximate nearest-neighbor search (ANNS) algorithms have become increasingly critical for recent AI applications, particularly in retrieval-augmented generation (RAG) and agent-based LLM applications. In this paper, we present CRINN, a new paradigm for ANNS algorithms. CRINN treats ANNS optimization as a reinforcement learning problem where execution speed serves as the reward signal. This approach enables the automatic generation of progressively faster ANNS implementations while maintaining accuracy constraints. Our experimental evaluation demonstrates CRINN's effectiveness across six widely-used NNS benchmark datasets. When compared against state-of-the-art open-source ANNS algorithms, CRINN achieves best performance on three of them (GIST-960-Euclidean, MNIST-784-Euclidean, and GloVe-25-angular), and tied for first place on two of them (SIFT-128-Euclidean and GloVe-25-angular). The implications of CRINN's success reach well beyond ANNS optimization: It validates that LLMs augmented with reinforcement learning can function as an effective tool for automating sophisticated algorithmic optimizations that demand specialized knowledge and labor-intensive manual refinement. Code can be found at this https URL
Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Databases (cs.DB)
arXiv:2508.02091 [cs.LG]
(or arXiv:2508.02091v4 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2508.02091 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 255 cuvinte
Subiecte conexe
Game Design Creativity evals
LessWrong · 12 oct. 2026
MURAL: Multimodal Uncertainty-aware Recommendation via Adaptive edge Learning
arXiv · 7 sept. 2026
Presentation: Postgres for Production Agents: Your Relational Foundation for Enterprise AI
InfoQ · 15 iul. 2026
Vector Search As Nearest Neighbor Matching: RAG-based Policy Learning in Causal Inference
arXiv · 21 iul. 2026