FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3204650139> ?p ?o ?g. }

• W3204650139 abstract "Question Answering (QA) has been a long-standing research topic in AI and NLP fields, and a wealth of studies have been conducted to attempt to equip QA systems with human-level reasoning capability. To approximate the complicated human reasoning process, state-of-the-art QA systems commonly use pre-trained language models (LMs) to access knowledge encoded in LMs together with elaborately designed modules based on Graph Neural Networks (GNNs) to perform reasoning over knowledge graphs (KGs). However, many problems remain open regarding the reasoning functionality of these GNN-based modules. Can these GNN-based modules really perform a complex reasoning process? Are they under- or over-complicated for QA? To open the black box of GNN and investigate these problems, we dissect state-of-the-art GNN modules for QA and analyze their reasoning capability. We discover that even a very simple graph neural counter can outperform all the existing GNN modules on CommonsenseQA and OpenBookQA, two popular QA benchmark datasets which heavily rely on knowledge-aware reasoning. Our work reveals that existing knowledge-aware GNN modules may only carry out some simple reasoning such as counting. It remains a challenging open problem to build comprehensive reasoning modules for knowledge-powered QA." @default.
• W3204650139 created "2021-10-11" @default.
• W3204650139 creator A5020025718 @default.
• W3204650139 creator A5030589527 @default.
• W3204650139 creator A5052596733 @default.
• W3204650139 creator A5076420985 @default.
• W3204650139 creator A5089413311 @default.
• W3204650139 date "2021-10-07" @default.
• W3204650139 modified "2023-10-11" @default.
• W3204650139 title "GNN is a Counter? Revisiting GNN for Question Answering" @default.
• W3204650139 cites W1793121960 @default.
• W3204650139 cites W1826234144 @default.
• W3204650139 cites W1909320841 @default.
• W3204650139 cites W2064675550 @default.
• W3204650139 cites W2094728533 @default.
• W3204650139 cites W2116341502 @default.
• W3204650139 cites W2166851633 @default.
• W3204650139 cites W2250539671 @default.
• W3204650139 cites W2251079237 @default.
• W3204650139 cites W2252136820 @default.
• W3204650139 cites W2283196293 @default.
• W3204650139 cites W2409591106 @default.
• W3204650139 cites W2519887557 @default.
• W3204650139 cites W2546950329 @default.
• W3204650139 cites W2582745083 @default.
• W3204650139 cites W2604314403 @default.
• W3204650139 cites W2624614404 @default.
• W3204650139 cites W2748428003 @default.
• W3204650139 cites W2752037867 @default.
• W3204650139 cites W2766453196 @default.
• W3204650139 cites W2785916643 @default.
• W3204650139 cites W2799238091 @default.
• W3204650139 cites W2890894339 @default.
• W3204650139 cites W2898695519 @default.
• W3204650139 cites W2900694899 @default.
• W3204650139 cites W2927302606 @default.
• W3204650139 cites W2949694638 @default.
• W3204650139 cites W2949800357 @default.
• W3204650139 cites W2949941638 @default.
• W3204650139 cites W2950339735 @default.
• W3204650139 cites W2950576363 @default.
• W3204650139 cites W2951004968 @default.
• W3204650139 cites W2963068946 @default.
• W3204650139 cites W2963341956 @default.
• W3204650139 cites W2963363373 @default.
• W3204650139 cites W2963403868 @default.
• W3204650139 cites W2963674932 @default.
• W3204650139 cites W2963895422 @default.
• W3204650139 cites W2964091467 @default.
• W3204650139 cites W2964207259 @default.
• W3204650139 cites W2964299589 @default.
• W3204650139 cites W2965373594 @default.
• W3204650139 cites W2971155257 @default.
• W3204650139 cites W2971869958 @default.
• W3204650139 cites W2972260442 @default.
• W3204650139 cites W2975059944 @default.
• W3204650139 cites W2983995706 @default.
• W3204650139 cites W2987669390 @default.
• W3204650139 cites W2994689640 @default.
• W3204650139 cites W2996848635 @default.
• W3204650139 cites W2998374885 @default.
• W3204650139 cites W3006188107 @default.
• W3204650139 cites W3006647218 @default.
• W3204650139 cites W3015440086 @default.
• W3204650139 cites W3021649351 @default.
• W3204650139 cites W3023160663 @default.
• W3204650139 cites W3023533951 @default.
• W3204650139 cites W3030163527 @default.
• W3204650139 cites W3035091181 @default.
• W3204650139 cites W3082274269 @default.
• W3204650139 cites W3099655892 @default.
• W3204650139 cites W3104178968 @default.
• W3204650139 cites W3104992282 @default.
• W3204650139 cites W3152801999 @default.
• W3204650139 doi "https://doi.org/10.48550/arxiv.2110.03192" @default.
• W3204650139 hasPublicationYear "2021" @default.
• W3204650139 type Work @default.
• W3204650139 sameAs 3204650139 @default.
• W3204650139 citedByCount "0" @default.
• W3204650139 crossrefType "posted-content" @default.
• W3204650139 hasAuthorship W3204650139A5020025718 @default.
• W3204650139 hasAuthorship W3204650139A5030589527 @default.
• W3204650139 hasAuthorship W3204650139A5052596733 @default.
• W3204650139 hasAuthorship W3204650139A5076420985 @default.
• W3204650139 hasAuthorship W3204650139A5089413311 @default.
• W3204650139 hasBestOaLocation W32046501391 @default.
• W3204650139 hasConcept C111472728 @default.
• W3204650139 hasConcept C132525143 @default.
• W3204650139 hasConcept C13280743 @default.
• W3204650139 hasConcept C138885662 @default.
• W3204650139 hasConcept C154945302 @default.
• W3204650139 hasConcept C161301231 @default.
• W3204650139 hasConcept C185798385 @default.
• W3204650139 hasConcept C199360897 @default.
• W3204650139 hasConcept C204321447 @default.
• W3204650139 hasConcept C205649164 @default.
• W3204650139 hasConcept C2780586882 @default.
• W3204650139 hasConcept C2987255567 @default.
• W3204650139 hasConcept C37335422 @default.
• W3204650139 hasConcept C41008148 @default.

• next