The Whole Truth and Nothing But the Truth: Faithful and Controllable Dialogue Response Generation...
681
17.5
Microsoft Research333 тыс
Следующее
07.07.23 – 82249:01Популярные
251 день – 6978:44
28.11.22 – 00:00Опубликовано 7 июля 2023, 18:47
The Whole Truth and Nothing But the Truth: Faithful and Controllable Dialogue Response Generation with Dataflow Transduction and Constrained Decoding
In a real-world dialogue system, generated text must be truthful and informative while remaining fluent and adhering to a prescribed style. Satisfying these constraints simultaneously is difficult for the two predominant paradigms in language generation: neural language modeling and rule-based generation. We describe a hybrid architecture for dialogue response generation that combines the strengths of both paradigms. The first component of this architecture is a rule-based content selection model defined using a new formal framework called dataflow transduction, which uses declarative rules to transduce a dialogue agent's actions and their results (represented as dataflow graphs) into context-free grammars representing the space of contextually acceptable responses. The second component is a constrained decoding procedure that uses these grammars to constrain the output of a neural language model, which selects fluent utterances. Our experiments show that this system outperforms both rule-based and learned approaches in human evaluations of fluency, relevance, and truthfulness.
See more at microsoft.com/en-us/research/v...
In a real-world dialogue system, generated text must be truthful and informative while remaining fluent and adhering to a prescribed style. Satisfying these constraints simultaneously is difficult for the two predominant paradigms in language generation: neural language modeling and rule-based generation. We describe a hybrid architecture for dialogue response generation that combines the strengths of both paradigms. The first component of this architecture is a rule-based content selection model defined using a new formal framework called dataflow transduction, which uses declarative rules to transduce a dialogue agent's actions and their results (represented as dataflow graphs) into context-free grammars representing the space of contextually acceptable responses. The second component is a constrained decoding procedure that uses these grammars to constrain the output of a neural language model, which selects fluent utterances. Our experiments show that this system outperforms both rule-based and learned approaches in human evaluations of fluency, relevance, and truthfulness.
See more at microsoft.com/en-us/research/v...
Свежие видео
8 дней – 40939:59
10 дней – 45 1810:42
13 дней – 1 3671:12
15 дней – 7714:56
16 дней – 8650:52
17 дней – 1 1350:40Случайные видео
333 дня – 441 5948:59
01.10.23 – 2 405 70737:37
19.07.23 – 6 8540:59
10.03.23 – 11 5962:13
03.02.22 – 16 64533:49
3 дня – 5 2600:49
5 дней – 49616:42
5 дней – 2 2870:26
7 дней – 925:26
9 дней – 27814:56
7 дней – 6356:27
7 дней – 574 5706:40