Triple-Encoders: Representations That Fire Together, Wire Together

• URL: http://arxiv.org/abs/2402.12332v2
• Date: Sat, 13 Jul 2024 17:58:25 GMT
• Title: Triple-Encoders: Representations That Fire Together, Wire Together
• Authors: Justus-Jonas Erker, Florian Mai, Nils Reimers, Gerasimos Spanakis, Iryna Gurevych,
• Abstract summary: Contrastive Learning is a representation learning method that encodes relative distances between utterances into the embedding space via a bi-encoder. This study introduces triple-encoders, which efficiently compute distributed utterance mixtures from these independently encoded utterances. We find that triple-encoders lead to a substantial improvement over bi-encoders, and even to better zero-shot generalization than single-vector representation models.
• Score: 51.15206713482718
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Search-based dialog models typically re-encode the dialog history at every turn, incurring high cost. Curved Contrastive Learning, a representation learning method that encodes relative distances between utterances into the embedding space via a bi-encoder, has recently shown promising results for dialog modeling at far superior efficiency. While high efficiency is achieved through independently encoding utterances, this ignores the importance of contextualization. To overcome this issue, this study introduces triple-encoders, which efficiently compute distributed utterance mixtures from these independently encoded utterances through a novel hebbian inspired co-occurrence learning objective in a self-organizing manner, without using any weights, i.e., merely through local interactions. Empirically, we find that triple-encoders lead to a substantial improvement over bi-encoders, and even to better zero-shot generalization than single-vector representation models without requiring re-encoding. Our code (https://github.com/UKPLab/acl2024-triple-encoders) and model (https://huggingface.co/UKPLab/triple-encoders-dailydialog) are publicly available.

Related papers

• Gesture2Text: A Generalizable Decoder for Word-Gesture Keyboards in XR Through Trajectory Coarse Discretization and Pre-training [2.81561528842917]
  We present a ready-to-use WGK decoder that is generalizable across mid-air and on-surface WGK systems in augmented reality (AR) and virtual reality (VR) It significantly outperforms SHARK2 with a 37.2% enhancement and surpasses the conventional neural decoder by 7.4%. It can operate in real-time, in just 97 milliseconds on Quest 3.
  arXiv  Detail & Related papers  (2024-10-08T12:53:22Z)
• Decoding at the Speed of Thought: Harnessing Parallel Decoding of Lexical Units for LLMs [57.27982780697922]
  Large language models have demonstrated exceptional capability in natural language understanding and generation. However, their generation speed is limited by the inherently sequential nature of their decoding process. This paper introduces Lexical Unit Decoding, a novel decoding methodology implemented in a data-driven manner.
  arXiv  Detail & Related papers  (2024-05-24T04:35:13Z)
• Sparse Autoencoders Enable Scalable and Reliable Circuit Identification in Language Models [0.0]
  This paper introduces an efficient and robust method for discovering interpretable circuits in large language models. We propose training sparse autoencoders on carefully designed positive and negative examples. Our findings highlight the promise of discrete sparse autoencoders for scalable and efficient mechanistic interpretability.
  arXiv  Detail & Related papers  (2024-05-21T06:26:10Z)
• Decoder-Only or Encoder-Decoder? Interpreting Language Model as a Regularized Encoder-Decoder [75.03283861464365]
  The seq2seq task aims at generating the target sequence based on the given input source sequence. Traditionally, most of the seq2seq task is resolved by an encoder to encode the source sequence and a decoder to generate the target text. Recently, a bunch of new approaches have emerged that apply decoder-only language models directly to the seq2seq task.
  arXiv  Detail & Related papers  (2023-04-08T15:44:29Z)
• Trans-Encoder: Unsupervised sentence-pair modelling through self- and mutual-distillations [22.40667024030858]
  Bi-encoders produce fixed-dimensional sentence representations and are computationally efficient. Cross-encoders can leverage their attention heads to exploit inter-sentence interactions for better performance. Trans-Encoder combines the two learning paradigms into an iterative joint framework to simultaneously learn enhanced bi- and cross-encoders.
  arXiv  Detail & Related papers  (2021-09-27T14:06:47Z)
• Uni-Encoder: A Fast and Accurate Response Selection Paradigm for Generation-Based Dialogue Systems [17.041715422600504]
  We develop a new encoding paradigm called Uni-Encoder. It keeps the full attention over each pair as in Cross-Encoder while only encoding the context once. It achieves new state-of-the-art results on four benchmark datasets with high computational efficiency.
  arXiv  Detail & Related papers  (2021-06-02T16:14:51Z)
• Rethinking Text Line Recognition Models [57.47147190119394]
  We consider two decoder families (Connectionist Temporal Classification and Transformer) and three encoder modules (Bidirectional LSTMs, Self-Attention, and GRCLs) We compare their accuracy and performance on widely used public datasets of scene and handwritten text. Unlike the more common Transformer-based models, this architecture can handle inputs of arbitrary length.
  arXiv  Detail & Related papers  (2021-04-15T21:43:13Z)
• Dual Encoding for Video Retrieval by Text [49.34356217787656]
  We propose a dual deep encoding network that encodes videos and queries into powerful dense representations of their own. Our novelty is two-fold. First, different from prior art that resorts to a specific single-level encoder, the proposed network performs multi-level encoding. Second, different from a conventional common space learning algorithm which is either concept based or latent space based, we introduce hybrid space learning.
  arXiv  Detail & Related papers  (2020-09-10T15:49:39Z)
• Learning Autoencoders with Relational Regularization [89.53065887608088]
  A new framework is proposed for learning autoencoders of data distributions. We minimize the discrepancy between the model and target distributions, with a emphrelational regularization We implement the framework with two scalable algorithms, making it applicable for both probabilistic and deterministic autoencoders.
  arXiv  Detail & Related papers  (2020-02-07T17:27:30Z)

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.