Pith. sign in

REVIEW

CANDLE: Decomposing Conditional and Conjunctive Queries for Task-Oriented Dialogue Systems

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 2107.03884 v2 pith:PSDZH5HX submitted 2021-07-08 cs.CL cs.AI

CANDLE: Decomposing Conditional and Conjunctive Queries for Task-Oriented Dialogue Systems

classification cs.CL cs.AI
keywords conditionalclassifiersdialoguequeriesactioncandlesequentialsingle
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Domain-specific dialogue systems generally determine user intents by relying on sentence level classifiers that mainly focus on single action sentences. Such classifiers are not designed to effectively handle complex queries composed of conditional and sequential clauses that represent multiple actions. We attempt to decompose such queries into smaller single action subqueries that are reasonable for intent classifiers to understand in a dialogue pipeline. We release, CANDLE(Conditional & AND type Expressions), a dataset consisting of 4282 utterances manually tagged with conditional and sequential labels, and demonstrates this decomposition by training two baseline taggers.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.