EHI: End-to-end Learning of Hierarchical Index for Efficient Dense Retrieval

• URL: http://arxiv.org/abs/2310.08891v2
• Date: Sun, 13 Oct 2024 04:49:45 GMT
• Title: EHI: End-to-end Learning of Hierarchical Index for Efficient Dense Retrieval
• Authors: Ramnath Kumar, Anshul Mittal, Nilesh Gupta, Aditya Kusupati, Inderjit Dhillon, Prateek Jain,
• Abstract summary: End-to-end Hierarchical Indexing (EHI) is a novel method for embedding-based retrieval. EHI outperforms existing state-of-the-art methods by +1.45% in MRR@10 on MS MARCO (Dev) and +8.2% in nDCG@10 on TREC DL19.
• Score: 18.15717995719973
• License:
• Abstract: Dense embedding-based retrieval is widely used for semantic search and ranking. However, conventional two-stage approaches, involving contrastive embedding learning followed by approximate nearest neighbor search (ANNS), can suffer from misalignment between these stages. This mismatch degrades retrieval performance. We propose End-to-end Hierarchical Indexing (EHI), a novel method that directly addresses this issue by jointly optimizing embedding generation and ANNS structure. EHI leverages a dual encoder for embedding queries and documents while simultaneously learning an inverted file index (IVF)-style tree structure. To facilitate the effective learning of this discrete structure, EHI introduces dense path embeddings that encodes the path traversed by queries and documents within the tree. Extensive evaluations on standard benchmarks, including MS MARCO (Dev set) and TREC DL19, demonstrate EHI's superiority over traditional ANNS index. Under the same computational constraints, EHI outperforms existing state-of-the-art methods by +1.45% in MRR@10 on MS MARCO (Dev) and +8.2% in nDCG@10 on TREC DL19, highlighting the benefits of our end-to-end approach.

Related papers

• Adaptive Retrieval and Scalable Indexing for k-NN Search with Cross-Encoders [77.84801537608651]
  Cross-encoder (CE) models which compute similarity by jointly encoding a query-item pair perform better than embedding-based models (dual-encoders) at estimating query-item relevance. We propose a sparse-matrix factorization based method that efficiently computes latent query and item embeddings to approximate CE scores and performs k-NN search with the approximate CE similarity.
  arXiv  Detail & Related papers  (2024-05-06T17:14:34Z)
• Semi-Parametric Retrieval via Binary Token Index [71.78109794895065]
  Semi-parametric Vocabulary Disentangled Retrieval (SVDR) is a novel semi-parametric retrieval framework. It supports two types of indexes: an embedding-based index for high effectiveness, akin to existing neural retrieval methods; and a binary token index that allows for quick and cost-effective setup, resembling traditional term-based retrieval. It achieves a 3% higher top-1 retrieval accuracy compared to the dense retriever DPR when using an embedding-based index and a 9% higher top-1 accuracy compared to BM25 when using a binary token index.
  arXiv  Detail & Related papers  (2024-05-03T08:34:13Z)
• Constructing Tree-based Index for Efficient and Effective Dense Retrieval [26.706985694158384]
  JTR stands for Joint optimization of TRee-based index and query encoding. We design a new unified contrastive learning loss to train tree-based index and query encoder in an end-to-end manner. Experimental results show that JTR achieves better retrieval performance while retaining high system efficiency.
  arXiv  Detail & Related papers  (2023-04-24T09:25:39Z)
• 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)
• End-to-End Learning to Index and Search in Large Output Spaces [95.16066833532396]
  Extreme multi-label classification (XMC) is a popular framework for solving real-world problems. In this paper, we propose a novel method which relaxes the tree-based index to a specialized weighted graph-based index. ELIAS achieves state-of-the-art performance on several large-scale extreme classification benchmarks with millions of labels.
  arXiv  Detail & Related papers  (2022-10-16T01:34:17Z)
• Hybrid Inverted Index Is a Robust Accelerator for Dense Retrieval [25.402767809863946]
  Inverted file structure is a common technique for accelerating dense retrieval. In this work, we present the Hybrid Inverted Index (HI$2$), where the embedding clusters and salient terms work to accelerate dense retrieval.
  arXiv  Detail & Related papers  (2022-10-11T15:12:41Z)
• Autoregressive Search Engines: Generating Substrings as Document Identifiers [53.0729058170278]
  Autoregressive language models are emerging as the de-facto standard for generating answers. Previous work has explored ways to partition the search space into hierarchical structures. In this work we propose an alternative that doesn't force any structure in the search space: using all ngrams in a passage as its possible identifiers.
  arXiv  Detail & Related papers  (2022-04-22T10:45:01Z)
• SPLADE v2: Sparse Lexical and Expansion Model for Information Retrieval [11.38022203865326]
  SPLADE model provides highly sparse representations and competitive results with respect to state-of-the-art dense and sparse approaches. We modify the pooling mechanism, benchmark a model solely based on document expansion, and introduce models trained with distillation. Overall, SPLADE is considerably improved with more than $9$% gains on NDCG@10 on TREC DL 2019, leading to state-of-the-art results on the BEIR benchmark.
  arXiv  Detail & Related papers  (2021-09-21T10:43:42Z)
• Micro-architectural Analysis of a Learned Index [0.0]
  ALEX is a tree-based in-memory index structure that consists of a hierarchy of machine learned models. Our results show that ALEX exhibits fewer stalls and a lower cycles-per-instruction value across different workloads. On the other hand, the amount of instructions required to handle out-of-bound inserts in ALEX can increase the instructions needed per request significantly (10X)
  arXiv  Detail & Related papers  (2021-09-17T12:13:06Z)
• Extracting Variable-Depth Logical Document Hierarchy from Long Documents: Method, Evaluation, and Application [21.270184491603864]
  We develop a framework, namely Hierarchy Extraction from Long Document (HELD), where we "sequentially" insert each physical object at the proper on of the current tree. Experiments based on thousands of long documents from Chinese, English financial market and English scientific publication. We show that logical document hierarchy can be employed to significantly improve the performance of the downstream passage retrieval task.
  arXiv  Detail & Related papers  (2021-05-14T06:26:22Z)
• Towards Improving the Consistency, Efficiency, and Flexibility of Differentiable Neural Architecture Search [84.4140192638394]
  Most differentiable neural architecture search methods construct a super-net for search and derive a target-net as its sub-graph for evaluation. In this paper, we introduce EnTranNAS that is composed of Engine-cells and Transit-cells. Our method also spares much memory and computation cost, which speeds up the search process.
  arXiv  Detail & Related papers  (2021-01-27T12:16:47Z)

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.