Controlled Text Generation as Continuous Optimization with Multiple Constraints

By Sachin Kumar, Eric Malmi, Aliaksei Severyn, Yulia Tsvetkov. 2021

90 citations recorded by the source corpus.

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.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Natural Language Processing

Citations in the corpus

Language Generation Models Can Cause Harm: So What Can We Do About It? An Actionable Survey (2022, methodology): “, 2022); Non-autoregressive factual error control (Kumar et al., 2021b)”
Generating Training Data with Language Models: Towards Zero-Shot Language Understanding (2022, background): “Adapting PLMs to generate texts of specific attributes can also be realized at inference time without any further training of the PLMs [10, 26, 27, 32, 47, 75].”
Tuning Language Models as Training Data Generators for Augmentation-Enhanced Few-Shot Learning (2022, background): “…phrases (Chan et al., 2021)) or both (Khal-ifa et al., 2021); at inference time, control over desired attributes can also be enforced without updating the PLM parameters (Dathathri et al., 2020; Krause et al., 2021; Kumar et al., 2021; Liu et al., 2021a; Pascual et al., 2021; Yang & Klein, 2021).”
Gradient-based Constrained Sampling from Language Models (2022, methodology): “After performing multiple steps of this gradient descent, discrete text can be extracted from ˜ y using different heuristics (Kumar et al., 2021; Qin et al., 2020; Song et al., 2020).”
COLD Decoding: Energy-based Constrained Text Generation with Langevin Dynamics (2022, unknown): “This “sample-and-select” approach differs from deterministic constrained decoding methods, which optimize only one sequence (e.g., Lu et al., 2021; Kumar et al., 2021), and is used widely in various generation settings (e.”
Mix and Match: Learning-free Controllable Text Generationusing Energy Language Models (2022, unknown): “Kumar et al. (2021) formulate the task of controlled generation as optimizing the base LM’s likelihood subject to global differentiable attribute-based constraints by gradient descent over the position-wise simplexes over the vocabulary.”