Controlled Text Generation as Continuous Optimization with Multiple Constraints

• URL: http://arxiv.org/abs/2108.01850v1
• Date: Wed, 4 Aug 2021 05:25:20 GMT
• Title: Controlled Text Generation as Continuous Optimization with Multiple Constraints
• Authors: Sachin Kumar, Eric Malmi, Aliaksei Severyn, Yulia Tsvetkov
• Abstract summary: We propose a flexible and modular algorithm for controllable inference from pretrained models. We make use of Lagrangian multipliers and gradient-descent based techniques to generate the desired text. We evaluate our approach on controllable machine translation and style transfer with multiple sentence-level attributes.
• Score: 23.71027518888138
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: As large-scale language model pretraining pushes the state-of-the-art in text generation, recent work has turned to controlling attributes of the text such models generate. While modifying the pretrained models via fine-tuning remains the popular approach, it incurs a significant computational cost and can be infeasible due to lack of appropriate data. As an alternative, we propose MuCoCO -- a flexible and modular algorithm for controllable inference from pretrained models. We formulate the decoding process as an optimization problem which allows for multiple attributes we aim to control to be easily incorporated as differentiable constraints to the optimization. By relaxing this discrete optimization to a continuous one, we make use of Lagrangian multipliers and gradient-descent based techniques to generate the desired text. We evaluate our approach on controllable machine translation and style transfer with multiple sentence-level attributes and observe significant improvements over baselines.

Related papers

• Style Transfer with Multi-iteration Preference Optimization [27.5647739554034]
  We consider the relationship between reinforcement learning and preference optimization. Inspired by these techniques from the past, we improve upon established preference optimization approaches. We evaluate our model on two commonly used text style transfer datasets.
  arXiv  Detail & Related papers  (2024-06-17T14:20:53Z)
• Model Ensembling for Constrained Optimization [7.4351710906830375]
  We consider a setting in which we wish to ensemble models for multidimensional output predictions that are in turn used for downstream optimization. More precisely, we imagine we are given a number of models mapping a state space to multidimensional real-valued predictions. These predictions form the coefficients of a linear objective that we would like to optimize under specified constraints. We apply multicalibration techniques that lead to two provably efficient and convergent algorithms.
  arXiv  Detail & Related papers  (2024-05-27T01:48:07Z)
• Controlled Text Generation via Language Model Arithmetic [7.687678490751105]
  We introduce model arithmetic, a novel inference framework for composing and biasing Large Language Models. We show that model arithmetic allows fine-grained control of generated text while outperforming state-of-the-art on the task of toxicity reduction.
  arXiv  Detail & Related papers  (2023-11-24T13:41:12Z)
• Amortizing intractable inference in large language models [56.92471123778389]
  We use amortized Bayesian inference to sample from intractable posterior distributions. We empirically demonstrate that this distribution-matching paradigm of LLM fine-tuning can serve as an effective alternative to maximum-likelihood training. As an important application, we interpret chain-of-thought reasoning as a latent variable modeling problem.
  arXiv  Detail & Related papers  (2023-10-06T16:36:08Z)
• When to Update Your Model: Constrained Model-based Reinforcement Learning [50.74369835934703]
  We propose a novel and general theoretical scheme for a non-decreasing performance guarantee of model-based RL (MBRL) Our follow-up derived bounds reveal the relationship between model shifts and performance improvement. A further example demonstrates that learning models from a dynamically-varying number of explorations benefit the eventual returns.
  arXiv  Detail & Related papers  (2022-10-15T17:57:43Z)
• A Template-based Method for Constrained Neural Machine Translation [100.02590022551718]
  We propose a template-based method that can yield results with high translation quality and match accuracy while keeping the decoding speed. The generation and derivation of the template can be learned through one sequence-to-sequence training framework. Experimental results show that the proposed template-based methods can outperform several representative baselines in lexically and structurally constrained translation tasks.
  arXiv  Detail & Related papers  (2022-05-23T12:24:34Z)
• SideControl: Controlled Open-domain Dialogue Generation via Additive Side Networks [10.607177634432214]
  We propose a novel approach to control the generation of Transformer-based pre-trained language models: the SideControl framework. Results show that the SideControl framework has better controllability, higher generation quality and better sample-efficiency than existing gradient-based and weighted-decoding baselines.
  arXiv  Detail & Related papers  (2021-09-05T01:15:26Z)
• GTAE: Graph-Transformer based Auto-Encoders for Linguistic-Constrained Text Style Transfer [119.70961704127157]
  Non-parallel text style transfer has attracted increasing research interests in recent years. Current approaches still lack the ability to preserve the content and even logic of original sentences. We propose a method called Graph Transformer based Auto-GTAE, which models a sentence as a linguistic graph and performs feature extraction and style transfer at the graph level.
  arXiv  Detail & Related papers  (2021-02-01T11:08:45Z)
• SDA: Improving Text Generation with Self Data Augmentation [88.24594090105899]
  We propose to improve the standard maximum likelihood estimation (MLE) paradigm by incorporating a self-imitation-learning phase for automatic data augmentation. Unlike most existing sentence-level augmentation strategies, our method is more general and could be easily adapted to any MLE-based training procedure.
  arXiv  Detail & Related papers  (2021-01-02T01:15:57Z)
• Unsupervised Paraphrasing with Pretrained Language Models [85.03373221588707]
  We propose a training pipeline that enables pre-trained language models to generate high-quality paraphrases in an unsupervised setting. Our recipe consists of task-adaptation, self-supervision, and a novel decoding algorithm named Dynamic Blocking. We show with automatic and human evaluations that our approach achieves state-of-the-art performance on both the Quora Question Pair and the ParaNMT datasets.
  arXiv  Detail & Related papers  (2020-10-24T11:55:28Z)
• Lexically Constrained Neural Machine Translation with Levenshtein Transformer [8.831954614241234]
  This paper proposes a simple and effective algorithm for incorporating lexical constraints in neural machine translation. Our method injects terminology constraints at inference time without any impact on decoding speed.
  arXiv  Detail & Related papers  (2020-04-27T09:59:27Z)

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.