Generating More Interesting Responses in Neural Conversation Models with Distributional Constraints
This addresses the issue of dull and repetitive outputs in conversational AI, making interactions more engaging for users, though it is an incremental improvement over existing methods.
The paper tackles the problem of neural conversation models generating safe, generic responses by proposing an approach that incorporates distributional constraints based on syntax, topics, and semantic similarity, resulting in responses that are much less generic without sacrificing plausibility, as shown through evaluation against competitive baselines.
Neural conversation models tend to generate safe, generic responses for most inputs. This is due to the limitations of likelihood-based decoding objectives in generation tasks with diverse outputs, such as conversation. To address this challenge, we propose a simple yet effective approach for incorporating side information in the form of distributional constraints over the generated responses. We propose two constraints that help generate more content rich responses that are based on a model of syntax and topics (Griffiths et al., 2005) and semantic similarity (Arora et al., 2016). We evaluate our approach against a variety of competitive baselines, using both automatic metrics and human judgments, showing that our proposed approach generates responses that are much less generic without sacrificing plausibility. A working demo of our code can be found at https://github.com/abaheti95/DC-NeuralConversation.