Pith. sign in

REVIEW

Contextual Fine-to-Coarse Distillation for Coarse-grained Response Selection in Open-Domain Conversations

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2109.13087 v2 pith:OHM7BJWR submitted 2021-09-24 cs.CL

classification cs.CL
keywords responsecoarse-grainedselectionarchitecturemodelcontextualconversationsdatasets
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We study the problem of coarse-grained response selection in retrieval-based dialogue systems. The problem is equally important with fine-grained response selection, but is less explored in existing literature. In this paper, we propose a Contextual Fine-to-Coarse (CFC) distilled model for coarse-grained response selection in open-domain conversations. In our CFC model, dense representations of query, candidate response and corresponding context is learned based on the multi-tower architecture, and more expressive knowledge learned from the one-tower architecture (fine-grained) is distilled into the multi-tower architecture (coarse-grained) to enhance the performance of the retriever. To evaluate the performance of our proposed model, we construct two new datasets based on the Reddit comments dump and Twitter corpus. Extensive experimental results on the two datasets show that the proposed methods achieve a significant improvement over all evaluation metrics compared with traditional baseline methods.

Discussion (0). Sign in to comment.

Pith tools