Matches in SemOpenAlex for { <https://semopenalex.org/work/W4289241457> ?p ?o ?g. }
Showing items 1 to 74 of
74
with 100 items per page.
- W4289241457 abstract "In several natural language tasks, labeled sequences are available in separate domains (say, languages), but the goal is to label sequences with mixed domain (such as code-switched text). Or, we may have available models for labeling whole passages (say, with sentiments), which we would like to exploit toward better position-specific label inference (say, target-dependent sentiment annotation). A key characteristic shared across such tasks is that different positions in a primary instance can benefit from different `experts' trained from auxiliary data, but labeled primary instances are scarce, and labeling the best expert for each position entails unacceptable cognitive burden. We propose GITNet, a unified position-sensitive multi-task recurrent neural network (RNN) architecture for such applications. Auxiliary and primary tasks need not share training instances. Auxiliary RNNs are trained over auxiliary instances. A primary instance is also submitted to each auxiliary RNN, but their state sequences are gated and merged into a novel composite state sequence tailored to the primary inference task. Our approach is in sharp contrast to recent multi-task networks like the cross-stitch and sluice network, which do not control state transfer at such fine granularity. We demonstrate the superiority of GIRNet using three applications: sentiment classification of code-switched passages, part-of-speech tagging of code-switched text, and target position-sensitive annotation of sentiment in monolingual passages. In all cases, we establish new state-of-the-art performance beyond recent competitive baselines." @default.
- W4289241457 created "2022-08-01" @default.
- W4289241457 creator A5008781655 @default.
- W4289241457 creator A5046521217 @default.
- W4289241457 creator A5082102505 @default.
- W4289241457 date "2018-11-28" @default.
- W4289241457 modified "2023-09-27" @default.
- W4289241457 title "GIRNet: Interleaved Multi-Task Recurrent State Sequence Models" @default.
- W4289241457 doi "https://doi.org/10.48550/arxiv.1811.11456" @default.
- W4289241457 hasPublicationYear "2018" @default.
- W4289241457 type Work @default.
- W4289241457 citedByCount "0" @default.
- W4289241457 crossrefType "posted-content" @default.
- W4289241457 hasAuthorship W4289241457A5008781655 @default.
- W4289241457 hasAuthorship W4289241457A5046521217 @default.
- W4289241457 hasAuthorship W4289241457A5082102505 @default.
- W4289241457 hasBestOaLocation W42892414571 @default.
- W4289241457 hasConcept C10138342 @default.
- W4289241457 hasConcept C119857082 @default.
- W4289241457 hasConcept C147168706 @default.
- W4289241457 hasConcept C154945302 @default.
- W4289241457 hasConcept C162324750 @default.
- W4289241457 hasConcept C177264268 @default.
- W4289241457 hasConcept C187736073 @default.
- W4289241457 hasConcept C198082294 @default.
- W4289241457 hasConcept C199360897 @default.
- W4289241457 hasConcept C204321447 @default.
- W4289241457 hasConcept C2776214188 @default.
- W4289241457 hasConcept C2776321320 @default.
- W4289241457 hasConcept C2776760102 @default.
- W4289241457 hasConcept C2778112365 @default.
- W4289241457 hasConcept C2780451532 @default.
- W4289241457 hasConcept C35639132 @default.
- W4289241457 hasConcept C41008148 @default.
- W4289241457 hasConcept C50644808 @default.
- W4289241457 hasConcept C54355233 @default.
- W4289241457 hasConcept C86803240 @default.
- W4289241457 hasConceptScore W4289241457C10138342 @default.
- W4289241457 hasConceptScore W4289241457C119857082 @default.
- W4289241457 hasConceptScore W4289241457C147168706 @default.
- W4289241457 hasConceptScore W4289241457C154945302 @default.
- W4289241457 hasConceptScore W4289241457C162324750 @default.
- W4289241457 hasConceptScore W4289241457C177264268 @default.
- W4289241457 hasConceptScore W4289241457C187736073 @default.
- W4289241457 hasConceptScore W4289241457C198082294 @default.
- W4289241457 hasConceptScore W4289241457C199360897 @default.
- W4289241457 hasConceptScore W4289241457C204321447 @default.
- W4289241457 hasConceptScore W4289241457C2776214188 @default.
- W4289241457 hasConceptScore W4289241457C2776321320 @default.
- W4289241457 hasConceptScore W4289241457C2776760102 @default.
- W4289241457 hasConceptScore W4289241457C2778112365 @default.
- W4289241457 hasConceptScore W4289241457C2780451532 @default.
- W4289241457 hasConceptScore W4289241457C35639132 @default.
- W4289241457 hasConceptScore W4289241457C41008148 @default.
- W4289241457 hasConceptScore W4289241457C50644808 @default.
- W4289241457 hasConceptScore W4289241457C54355233 @default.
- W4289241457 hasConceptScore W4289241457C86803240 @default.
- W4289241457 hasLocation W42892414571 @default.
- W4289241457 hasLocation W42892414572 @default.
- W4289241457 hasOpenAccess W4289241457 @default.
- W4289241457 hasPrimaryLocation W42892414571 @default.
- W4289241457 hasRelatedWork W2227316993 @default.
- W4289241457 hasRelatedWork W2834136616 @default.
- W4289241457 hasRelatedWork W2894887961 @default.
- W4289241457 hasRelatedWork W3116646283 @default.
- W4289241457 hasRelatedWork W3164092048 @default.
- W4289241457 hasRelatedWork W4205820553 @default.
- W4289241457 hasRelatedWork W4281386417 @default.
- W4289241457 hasRelatedWork W4322096459 @default.
- W4289241457 hasRelatedWork W4327831767 @default.
- W4289241457 hasRelatedWork W54903681 @default.
- W4289241457 isParatext "false" @default.
- W4289241457 isRetracted "false" @default.
- W4289241457 workType "article" @default.