NeuRoute: Logit-Guided Neural Routing for Billion-Scale Vector Search with Sub-Hour Index Construction

• URL: http://arxiv.org/abs/2608.15438v1
• Date: Sat, 15 Aug 2026 22:44:10 GMT
• Title: NeuRoute: Logit-Guided Neural Routing for Billion-Scale Vector Search with Sub-Hour Index Construction
• Abstract summary: NeuRoute is a learned hashing index that turns short binary codes into an effective routing primitive for large-scale vector search.<n>NeuRoute trains a lightweight neural network encoder with a selective similarity-preserving objective to produce well-balanced binary addresses.
• Score: 6.022622976208937
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Building approximate nearest neighbor (ANN) indexes at billion scale is often dominated by expensive global clustering or graph construction, making time-to-index a first-order systems concern. We present NeuRoute, a learned hashing index that turns short binary codes into an effective routing primitive for large-scale vector search. NeuRoute trains a lightweight neural network encoder with a selective similarity-preserving objective to produce well-balanced binary addresses. During construction, NeuRoute organizes vectors into buckets by their codes and performs bucket-local clustering in the encoder's low-dimensional space to form centroids. At query time, NeuRoute exploits the encoder logits as an uncertainty signal: it uses deviation-to-threshold scores to prioritize uncertain-bit perturbations for query-adaptive multi-bucket probing, scores bucket-local centroids by their distances to the query to form a compact candidate cluster set, and applies centroid-stage gating with heap-quality-driven early stopping to prune low-value clusters before exact refinement. On billion-scale benchmarks, NeuRoute achieves strong accuracy-throughput trade-offs with fast index construction: on BigANN-1B it reaches $90.3\%$ Recall@10 at 2,414 QPS and is $1.7\times$ faster than OPQ+IVF-PQ (refine) at comparable accuracy, while completing end-to-end training+construction in under an hour on both BigANN-1B and Deep1B-1B. These results show that logit-guided neural routing can make hashing competitive as a lightweight ANN indexing framework at billion scale. Source code and artifacts are available at https://github.com/XingqiaoWang/NeuRoute.

Related papers

• Efficient and Trainable Language Model Test-Time Scaling via Local Branch Routing [65.59464371673623]
  Local Branch Routing (LBR) is a token-level test-time scaling framework.<n>It expands a small local lookahead tree, forwards all sampled branches through the language model, and uses a lightweight router to select the depth-1 subtree to commit.<n>LBR allows each token decision to use evidence beyond the root next-token distribution while avoiding full solution-level search.
  arXiv  Detail & Related papers  (2026-06-24T03:42:44Z)
• PiPNN: Ultra-Scalable Graph-Based Nearest Neighbor Indexing [6.243421401579046]
  PiPNN is an ultra-scalable graph construction algorithm that avoids the search bottleneck'' that existing graph-based methods suffer from.<n>HashPrune enables PiPNN to partition the dataset into overlapping sub-problems, efficiently perform bulk distance comparisons, and stream a subset of the edges into HashPrune.<n>PiPNN builds state-of-the-art indexes up to 11.6x faster than Vamana and up to 12.9x faster than HNSW.
  arXiv  Detail & Related papers  (2026-02-17T02:18:17Z)
• OrchANN: A Unified I/O Orchestration Framework for Skewed Out-of-Core Vector Search [38.2797436367431]
  Approximate nearest neighbor search (ANNS) at billion scale is fundamentally an out-of-core problem.<n>We present OrchANN, an out-of-core ANNS engine that uses an I/O orchestration model for unified I/O governance.
  arXiv  Detail & Related papers  (2025-12-28T08:42:38Z)
• Efficient k-NN Search with Cross-Encoders using Adaptive Multi-Round CUR Decomposition [77.4863142882136]
  Cross-encoder models are prohibitively expensive for direct k-nearest neighbor (k-NN) search. We propose ADACUR, a method that adaptively, iteratively, and efficiently minimizes the approximation error for the practically important top-k neighbors.
  arXiv  Detail & Related papers  (2023-05-04T17:01:17Z)
• Improving Dual-Encoder Training through Dynamic Indexes for Negative Mining [61.09807522366773]
  We introduce an algorithm that approximates the softmax with provable bounds and that dynamically maintains the tree. In our study on datasets with over twenty million targets, our approach cuts error by half in relation to oracle brute-force negative mining.
  arXiv  Detail & Related papers  (2023-03-27T15:18:32Z)
• Neural Architecture Search: Two Constant Shared Weights Initialisations [0.0]
  epsinas is a novel zero-cost NAS metric that assesses architecture potential using two constant shared weight initialisations and the statistics of their outputs.<n>We show that the dispersion of raw outputs, normalised by their average magnitude, strongly correlates with trained accuracy.<n>Our computation requires no data labels, operates on a single minibatch, and eliminates the need for gradient.
  arXiv  Detail & Related papers  (2023-02-09T02:25:38Z)
• Rapid Person Re-Identification via Sub-space Consistency Regularization [51.76876061721556]
  Person Re-Identification (ReID) matches pedestrians across disjoint cameras. Existing ReID methods adopting real-value feature descriptors have achieved high accuracy, but they are low in efficiency due to the slow Euclidean distance computation. We propose a novel Sub-space Consistency Regularization (SCR) algorithm that can speed up the ReID procedure by 0.25$ times.
  arXiv  Detail & Related papers  (2022-07-13T02:44:05Z)
• Neural Architecture Search via Bregman Iterations [1.7778609937758323]
  We propose a novel strategy for Neural Architecture Search (NAS) based on Bregman iterations. Our gradient-based one-shot algorithm gradually adds relevant parameters in an inverse scale space manner.
  arXiv  Detail & Related papers  (2021-06-04T13:37:47Z)
• IRLI: Iterative Re-partitioning for Learning to Index [104.72641345738425]
  Methods have to trade between obtaining high accuracy while maintaining load balance and scalability in distributed settings. We propose a novel approach called IRLI, which iteratively partitions the items by learning the relevant buckets directly from the query-item relevance data. We mathematically show that IRLI retrieves the correct item with high probability under very natural assumptions and provides superior load balancing.
  arXiv  Detail & Related papers  (2021-03-17T23:13:25Z)
• Cluster-and-Conquer: When Randomness Meets Graph Locality [0.0]
  Some of the most efficient KNN graph algorithms are incremental and local. Cluster-and-Conquer boosts the starting configuration of greedy algorithms.
  arXiv  Detail & Related papers  (2020-10-22T07:31:12Z)
• Learning to Accelerate Heuristic Searching for Large-Scale Maximum Weighted b-Matching Problems in Online Advertising [51.97494906131859]
  Bipartite b-matching is fundamental in algorithm design, and has been widely applied into economic markets, labor markets, etc. Existing exact and approximate algorithms usually fail in such settings due to either requiring intolerable running time or too much computation resource. We propose textttNeuSearcher which leverages the knowledge learned from previously instances to solve new problem instances.
  arXiv  Detail & Related papers  (2020-05-09T02:48:23Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.