Neural Rule-Execution Tracking Machine For Transformer-Based Text Generation
This addresses the need for scalable and unified rule execution in text generation for applications requiring control and prior knowledge, representing a novel method rather than an incremental improvement.
The paper tackles the problem of executing multiple specific rules (e.g., controllable constraints) in transformer-based text generation models, which are typically black-box and require case-by-case designs, by proposing a Neural Rule-Execution Tracking Machine that enables simultaneous rule guidance, leading to superior performance verified on several benchmarks.
Sequence-to-Sequence (S2S) neural text generation models, especially the pre-trained ones (e.g., BART and T5), have exhibited compelling performance on various natural language generation tasks. However, the black-box nature of these models limits their application in tasks where specific rules (e.g., controllable constraints, prior knowledge) need to be executed. Previous works either design specific model structure (e.g., Copy Mechanism corresponding to the rule "the generated output should include certain words in the source input") or implement specialized inference algorithm (e.g., Constrained Beam Search) to execute particular rules through the text generation. These methods require careful design case-by-case and are difficult to support multiple rules concurrently. In this paper, we propose a novel module named Neural Rule-Execution Tracking Machine that can be equipped into various transformer-based generators to leverage multiple rules simultaneously to guide the neural generation model for superior generation performance in a unified and scalable way. Extensive experimental results on several benchmarks verify the effectiveness of our proposed model in both controllable and general text generation.