To Asymmetry and Beyond: Structured Pruning of Sequence to Sequence Models for Improved Inference Efficiency

• URL: http://arxiv.org/abs/2304.02721v3
• Date: Mon, 12 Jun 2023 21:13:14 GMT
• Title: To Asymmetry and Beyond: Structured Pruning of Sequence to Sequence Models for Improved Inference Efficiency
• Authors: Daniel Campos, ChengXiang Zhai
• Abstract summary: We show that model accuracy is tied to the encoder size while inference efficiency is connected to the decoder. We find both the average degradation and the role of asymmetry to be consistent across model sizes and variations in datasets.
• Score: 37.22592489907125
• License: http://creativecommons.org/publicdomain/zero/1.0/
• Abstract: Sequence-to-sequence language models can be used to produce abstractive summaries which are coherent, relevant, and concise. Still, model sizes can make deployment in latency-sensitive or web-scale implementations difficult. This paper studies the relationship between model size, structured pruning, inference efficiency, and summarization accuracy on widely used summarization datasets. We show that model accuracy is tied to the encoder size while inference efficiency is connected to the decoder. Using asymmetric pruning can lead to nearly 3x improvement in inference latency with ~1 point loss in Rouge-2. Moreover, we find both the average degradation and the role of asymmetry to be consistent across model sizes and variations in datasets.

Related papers

• Combining Local Symmetry Exploitation and Reinforcement Learning for Optimised Probabilistic Inference -- A Work In Progress [2.2164989053903805]
  Efficient probabilistic inference by variable elimination in graphical models requires an optimal elimination order. We adapt a reinforcement learning approach to find efficient contraction orders in tensor networks. We show that leveraging specific structures during inference allows for introducing compact encodings of intermediate results.
  arXiv  Detail & Related papers  (2025-03-11T18:00:23Z)
• Revisiting Cascaded Ensembles for Efficient Inference [32.914852531806]
  A common approach to make machine learning inference more efficient is to use example-specific adaptive schemes. In this work we study a simple scheme for adaptive inference. We build a cascade of ensembles (CoE), beginning with resource-efficient models and growing to larger, more expressive models.
  arXiv  Detail & Related papers  (2024-07-02T15:14:12Z)
• Calibrating Likelihoods towards Consistency in Summarization Models [22.023863165579602]
  We argue that the main reason for such behavior is that the summarization models trained with maximum likelihood objective assign high probability to plausible sequences given the context. In this work, we solve this problem by calibrating the likelihood of model generated sequences to better align with a consistency metric measured by natural language inference (NLI) models.
  arXiv  Detail & Related papers  (2023-10-12T23:17:56Z)
• Variable Importance Matching for Causal Inference [73.25504313552516]
  We describe a general framework called Model-to-Match that achieves these goals. Model-to-Match uses variable importance measurements to construct a distance metric. We operationalize the Model-to-Match framework with LASSO.
  arXiv  Detail & Related papers  (2023-02-23T00:43:03Z)
• Sketching as a Tool for Understanding and Accelerating Self-attention for Long Sequences [52.6022911513076]
  Transformer-based models are not efficient in processing long sequences due to the quadratic space and time complexity of the self-attention modules. We propose Linformer and Informer to reduce the quadratic complexity to linear (modulo logarithmic factors) via low-dimensional projection and row selection. Based on the theoretical analysis, we propose Skeinformer to accelerate self-attention and further improve the accuracy of matrix approximation to self-attention.
  arXiv  Detail & Related papers  (2021-12-10T06:58:05Z)
• Recursive Contour Saliency Blending Network for Accurate Salient Object Detection [0.0]
  In this work, we designed a network for better edge quality in salient object detection. We proposed a contour-saliency blending module to exchange information between contour and saliency. Our model is lightweight and fast, with only 27.9 million parameters and real-time inferencing at 31 FPS.
  arXiv  Detail & Related papers  (2021-05-28T14:19:54Z)
• Stacking VAE with Graph Neural Networks for Effective and Interpretable Time Series Anomaly Detection [5.935707085640394]
  We propose a stacking variational auto-encoder (VAE) model with graph neural networks for the effective and interpretable time-series anomaly detection. We show that our proposed model outperforms the strong baselines on three public datasets with considerable improvements.
  arXiv  Detail & Related papers  (2021-05-18T09:50:00Z)
• Anomaly Detection of Time Series with Smoothness-Inducing Sequential Variational Auto-Encoder [59.69303945834122]
  We present a Smoothness-Inducing Sequential Variational Auto-Encoder (SISVAE) model for robust estimation and anomaly detection of time series. Our model parameterizes mean and variance for each time-stamp with flexible neural networks. We show the effectiveness of our model on both synthetic datasets and public real-world benchmarks.
  arXiv  Detail & Related papers  (2021-02-02T06:15:15Z)
• Slice Sampling for General Completely Random Measures [74.24975039689893]
  We present a novel Markov chain Monte Carlo algorithm for posterior inference that adaptively sets the truncation level using auxiliary slice variables. The efficacy of the proposed algorithm is evaluated on several popular nonparametric models.
  arXiv  Detail & Related papers  (2020-06-24T17:53:53Z)
• Triple Wins: Boosting Accuracy, Robustness and Efficiency Together by Enabling Input-Adaptive Inference [119.19779637025444]
  Deep networks were recently suggested to face the odds between accuracy (on clean natural images) and robustness (on adversarially perturbed images) This paper studies multi-exit networks associated with input-adaptive inference, showing their strong promise in achieving a "sweet point" in cooptimizing model accuracy, robustness and efficiency.
  arXiv  Detail & Related papers  (2020-02-24T00:40:22Z)

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.