Semantic Space Grounded Weighted Decoding for Multi-Attribute Controllable Dialogue Generation
This addresses a practically useful but under-studied problem for chatbot developers, though it appears incremental as it builds on weighted decoding paradigms.
The paper tackled the problem of controlling chatbot utterance generation with multiple attributes like personalities, emotions, and dialogue acts, proposing the DASC framework that achieves high control accuracy with simultaneous control of 3 aspects while reducing model sizes.
Controlling chatbot utterance generation with multiple attributes such as personalities, emotions and dialogue acts is a practically useful but under-studied problem. We propose a novel framework called DASC that possesses strong controllability with a weighted decoding paradigm, while improving generation quality with the grounding in an attribute semantics space. Generation with multiple attributes is then intuitively implemented with an interpolation of multiple attribute embeddings, which results in substantial reduction in the model sizes. Experiments show that DASC can achieve high control accuracy in generation task with the simultaneous control of 3 aspects while also producing interesting and reasonably sensible responses, even in an out-of-distribution robustness test.