Matches in SemOpenAlex for { <https://semopenalex.org/work/W3213016591> ?p ?o ?g. }
- W3213016591 abstract "Despite the success of neural dialogue systems in achieving high performance on the leader-board, they cannot meet users' requirements in practice, due to their poor reasoning skills. The underlying reason is that most neural dialogue models only capture the syntactic and semantic information, but fail to model the logical consistency between the dialogue history and the generated response. Recently, a new multi-turn dialogue reasoning task has been proposed, to facilitate dialogue reasoning research. However, this task is challenging, because there are only slight differences between the illogical response and the dialogue history. How to effectively solve this challenge is still worth exploring. This paper proposes a Fine-grained Comparison Model (FCM) to tackle this problem. Inspired by human's behavior in reading comprehension, a comparison mechanism is proposed to focus on the fine-grained differences in the representation of each response candidate. Specifically, each candidate representation is compared with the whole history to obtain a history consistency representation. Furthermore, the consistency signals between each candidate and the speaker's own history are considered to drive a model to prefer a candidate that is logically consistent with the speaker's history logic. Finally, the above consistency representations are employed to output a ranking list of the candidate responses for multi-turn dialogue reasoning. Experimental results on two public dialogue datasets show that our method obtains higher ranking scores than the baseline models." @default.
- W3213016591 created "2021-11-22" @default.
- W3213016591 creator A5008043408 @default.
- W3213016591 creator A5020444844 @default.
- W3213016591 creator A5032087333 @default.
- W3213016591 creator A5035164990 @default.
- W3213016591 creator A5041432735 @default.
- W3213016591 creator A5056101043 @default.
- W3213016591 creator A5066907634 @default.
- W3213016591 creator A5077926417 @default.
- W3213016591 date "2021-09-22" @default.
- W3213016591 modified "2023-09-25" @default.
- W3213016591 title "FCM: A Fine-grained Comparison Model for Multi-turn Dialogue Reasoning" @default.
- W3213016591 cites W102065888 @default.
- W3213016591 cites W2009593660 @default.
- W3213016591 cites W2581637843 @default.
- W3213016591 cites W2789033601 @default.
- W3213016591 cites W2789078277 @default.
- W3213016591 cites W2798456655 @default.
- W3213016591 cites W2807791032 @default.
- W3213016591 cites W2862781886 @default.
- W3213016591 cites W2891416139 @default.
- W3213016591 cites W2908285010 @default.
- W3213016591 cites W2947812138 @default.
- W3213016591 cites W2952813980 @default.
- W3213016591 cites W2962854379 @default.
- W3213016591 cites W2962883855 @default.
- W3213016591 cites W2962896208 @default.
- W3213016591 cites W2963341956 @default.
- W3213016591 cites W2963520511 @default.
- W3213016591 cites W2963544536 @default.
- W3213016591 cites W2963986868 @default.
- W3213016591 cites W2964121744 @default.
- W3213016591 cites W2965373594 @default.
- W3213016591 cites W2970648534 @default.
- W3213016591 cites W2972324944 @default.
- W3213016591 cites W3034446185 @default.
- W3213016591 cites W3035356453 @default.
- W3213016591 cites W3094612274 @default.
- W3213016591 cites W3101227179 @default.
- W3213016591 cites W3121541553 @default.
- W3213016591 cites W3172448816 @default.
- W3213016591 doi "https://doi.org/10.48550/arxiv.2109.10510" @default.
- W3213016591 hasPublicationYear "2021" @default.
- W3213016591 type Work @default.
- W3213016591 sameAs 3213016591 @default.
- W3213016591 citedByCount "0" @default.
- W3213016591 crossrefType "posted-content" @default.
- W3213016591 hasAuthorship W3213016591A5008043408 @default.
- W3213016591 hasAuthorship W3213016591A5020444844 @default.
- W3213016591 hasAuthorship W3213016591A5032087333 @default.
- W3213016591 hasAuthorship W3213016591A5035164990 @default.
- W3213016591 hasAuthorship W3213016591A5041432735 @default.
- W3213016591 hasAuthorship W3213016591A5056101043 @default.
- W3213016591 hasAuthorship W3213016591A5066907634 @default.
- W3213016591 hasAuthorship W3213016591A5077926417 @default.
- W3213016591 hasBestOaLocation W32130165911 @default.
- W3213016591 hasConcept C119857082 @default.
- W3213016591 hasConcept C120665830 @default.
- W3213016591 hasConcept C121332964 @default.
- W3213016591 hasConcept C154945302 @default.
- W3213016591 hasConcept C162324750 @default.
- W3213016591 hasConcept C17744445 @default.
- W3213016591 hasConcept C187736073 @default.
- W3213016591 hasConcept C189430467 @default.
- W3213016591 hasConcept C192209626 @default.
- W3213016591 hasConcept C199360897 @default.
- W3213016591 hasConcept C199539241 @default.
- W3213016591 hasConcept C204321447 @default.
- W3213016591 hasConcept C2776359362 @default.
- W3213016591 hasConcept C2776436953 @default.
- W3213016591 hasConcept C2780451532 @default.
- W3213016591 hasConcept C41008148 @default.
- W3213016591 hasConcept C511192102 @default.
- W3213016591 hasConcept C94625758 @default.
- W3213016591 hasConceptScore W3213016591C119857082 @default.
- W3213016591 hasConceptScore W3213016591C120665830 @default.
- W3213016591 hasConceptScore W3213016591C121332964 @default.
- W3213016591 hasConceptScore W3213016591C154945302 @default.
- W3213016591 hasConceptScore W3213016591C162324750 @default.
- W3213016591 hasConceptScore W3213016591C17744445 @default.
- W3213016591 hasConceptScore W3213016591C187736073 @default.
- W3213016591 hasConceptScore W3213016591C189430467 @default.
- W3213016591 hasConceptScore W3213016591C192209626 @default.
- W3213016591 hasConceptScore W3213016591C199360897 @default.
- W3213016591 hasConceptScore W3213016591C199539241 @default.
- W3213016591 hasConceptScore W3213016591C204321447 @default.
- W3213016591 hasConceptScore W3213016591C2776359362 @default.
- W3213016591 hasConceptScore W3213016591C2776436953 @default.
- W3213016591 hasConceptScore W3213016591C2780451532 @default.
- W3213016591 hasConceptScore W3213016591C41008148 @default.
- W3213016591 hasConceptScore W3213016591C511192102 @default.
- W3213016591 hasConceptScore W3213016591C94625758 @default.
- W3213016591 hasLocation W32130165911 @default.
- W3213016591 hasOpenAccess W3213016591 @default.
- W3213016591 hasPrimaryLocation W32130165911 @default.
- W3213016591 hasRelatedWork W1546381263 @default.
- W3213016591 hasRelatedWork W2353865532 @default.
- W3213016591 hasRelatedWork W2961085424 @default.
- W3213016591 hasRelatedWork W3046775127 @default.