Hypencoder: Hypernetworks for Information Retrieval

• URL: http://arxiv.org/abs/2502.05364v2
• Date: Thu, 01 May 2025 16:43:27 GMT
• Title: Hypencoder: Hypernetworks for Information Retrieval
• Authors: Julian Killingback, Hansi Zeng, Hamed Zamani,
• Abstract summary: We use a small neural network that acts as a learned query-specific relevance function.<n>We show that our model is able to retrieve from a corpus of 8.8M documents in under 60 milliseconds.
• Score: 20.173669986209024
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Existing information retrieval systems are largely constrained by their reliance on vector inner products to assess query-document relevance, which naturally limits the expressiveness of the relevance score they can produce. We propose a new paradigm; instead of representing a query as a vector, we use a small neural network that acts as a learned query-specific relevance function. This small neural network takes a document representation as input (in this work we use a single vector) and produces a scalar relevance score. To produce the small neural network we use a hypernetwork, a network that produces the weights of other networks, as our query encoder. We name this category of encoder models Hypencoders. Experiments on in-domain search tasks show that Hypencoders significantly outperform strong dense retrieval models and even surpass reranking models and retrieval models with an order of magnitude more parameters. To assess the extent of Hypencoders' capabilities, we evaluate on a set of hard retrieval tasks including tip-of-the-tongue and instruction-following retrieval tasks. On harder tasks, we find that the performance gap widens substantially compared to standard retrieval tasks. Furthermore, to demonstrate the practicality of our method, we implement an approximate search algorithm and show that our model is able to retrieve from a corpus of 8.8M documents in under 60 milliseconds.

Related papers

• A Fresh Take on Stale Embeddings: Improving Dense Retriever Training with Corrector Networks [81.2624272756733]
  In dense retrieval, deep encoders provide embeddings for both inputs and targets. We train a small parametric corrector network that adjusts stale cached target embeddings. Our approach matches state-of-the-art results even when no target embedding updates are made during training.
  arXiv  Detail & Related papers  (2024-09-03T13:29:13Z)
• Towards Scalable and Versatile Weight Space Learning [51.78426981947659]
  This paper introduces the SANE approach to weight-space learning. Our method extends the idea of hyper-representations towards sequential processing of subsets of neural network weights.
  arXiv  Detail & Related papers  (2024-06-14T13:12:07Z)
• How Does Generative Retrieval Scale to Millions of Passages? [68.98628807288972]
  We conduct the first empirical study of generative retrieval techniques across various corpus scales. We scale generative retrieval to millions of passages with a corpus of 8.8M passages and evaluating model sizes up to 11B parameters. While generative retrieval is competitive with state-of-the-art dual encoders on small corpora, scaling to millions of passages remains an important and unsolved challenge.
  arXiv  Detail & Related papers  (2023-05-19T17:33:38Z)
• Tram: A Token-level Retrieval-augmented Mechanism for Source Code Summarization [76.57699934689468]
  We propose a fine-grained Token-level retrieval-augmented mechanism (Tram) on the decoder side to enhance the performance of neural models. To overcome the challenge of token-level retrieval in capturing contextual code semantics, we also propose integrating code semantics into individual summary tokens.
  arXiv  Detail & Related papers  (2023-05-18T16:02:04Z)
• CorpusBrain: Pre-train a Generative Retrieval Model for Knowledge-Intensive Language Tasks [62.22920673080208]
  Single-step generative model can dramatically simplify the search process and be optimized in end-to-end manner. We name the pre-trained generative retrieval model as CorpusBrain as all information about the corpus is encoded in its parameters without the need of constructing additional index.
  arXiv  Detail & Related papers  (2022-08-16T10:22:49Z)
• GROWN+UP: A Graph Representation Of a Webpage Network Utilizing Pre-training [0.2538209532048866]
  We introduce an agnostic deep graph neural network feature extractor that can ingest webpage structures, pre-train self-supervised on massive unlabeled data, and fine-tune to arbitrary tasks on webpages effectually. We show that our pre-trained model achieves state-of-the-art results using multiple datasets on two very different benchmarks: webpage boilerplate removal and genre classification.
  arXiv  Detail & Related papers  (2022-08-03T13:37:27Z)
• Discrete Key-Value Bottleneck [95.61236311369821]
  Deep neural networks perform well on classification tasks where data streams are i.i.d. and labeled data is abundant. One powerful approach that has addressed this challenge involves pre-training of large encoders on volumes of readily available data, followed by task-specific tuning. Given a new task, however, updating the weights of these encoders is challenging as a large number of weights needs to be fine-tuned, and as a result, they forget information about the previous tasks. We propose a model architecture to address this issue, building upon a discrete bottleneck containing pairs of separate and learnable key-value codes.
  arXiv  Detail & Related papers  (2022-07-22T17:52:30Z)
• Variable Bitrate Neural Fields [75.24672452527795]
  We present a dictionary method for compressing feature grids, reducing their memory consumption by up to 100x. We formulate the dictionary optimization as a vector-quantized auto-decoder problem which lets us learn end-to-end discrete neural representations in a space where no direct supervision is available.
  arXiv  Detail & Related papers  (2022-06-15T17:58:34Z)
• Gluing Neural Networks Symbolically Through Hyperdimensional Computing [8.209945970790741]
  We explore the notion of using binary hypervectors to encode the final, classifying output signals of neural networks. This allows multiple neural networks to work together to solve a problem, with little additional overhead. We find that this outperforms the state of the art, or is on a par with it, while using very little overhead.
  arXiv  Detail & Related papers  (2022-05-31T04:44:02Z)
• ED2LM: Encoder-Decoder to Language Model for Faster Document Re-ranking Inference [70.36083572306839]
  This paper proposes a new training and inference paradigm for re-ranking. We finetune a pretrained encoder-decoder model using in the form of document to query generation. We show that this encoder-decoder architecture can be decomposed into a decoder-only language model during inference.
  arXiv  Detail & Related papers  (2022-04-25T06:26:29Z)
• 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)
• Learning Deep Semantic Model for Code Search using CodeSearchNet Corpus [17.6095840480926]
  We propose a novel deep semantic model which makes use of the utilities of multi-modal sources. We apply the proposed model to tackle the CodeSearchNet challenge about semantic code search. Our model is trained on CodeSearchNet corpus and evaluated on the held-out data, the final model achieves 0.384 NDCG and won the first place in this benchmark.
  arXiv  Detail & Related papers  (2022-01-27T04:15:59Z)
• A Variational Graph Autoencoder for Manipulation Action Recognition and Prediction [1.1816942730023883]
  We introduce a deep graph autoencoder to jointly learn recognition and prediction of manipulation tasks from symbolic scene graphs. Our network has a variational autoencoder structure with two branches: one for identifying the input graph type and one for predicting the future graphs. We benchmark our new model against different state-of-the-art methods on two different datasets, MANIAC and MSRC-9, and show that our proposed model can achieve better performance.
  arXiv  Detail & Related papers  (2021-10-25T21:40:42Z)
• Learning Purified Feature Representations from Task-irrelevant Labels [18.967445416679624]
  We propose a novel learning framework called PurifiedLearning to exploit task-irrelevant features extracted from task-irrelevant labels. Our work is built on solid theoretical analysis and extensive experiments, which demonstrate the effectiveness of PurifiedLearning.
  arXiv  Detail & Related papers  (2021-02-22T12:50:49Z)
• Sparse, Dense, and Attentional Representations for Text Retrieval [25.670835450331943]
  Dual encoders perform retrieval by encoding documents and queries into dense lowdimensional vectors. We investigate the capacity of this architecture relative to sparse bag-of-words models and attentional neural networks. We propose a simple neural model that combines the efficiency of dual encoders with some of the expressiveness of more costly attentional architectures.
  arXiv  Detail & Related papers  (2020-05-01T02:21:17Z)

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.