Matches in SemOpenAlex for { <https://semopenalex.org/work/W3170842411> ?p ?o ?g. }
Showing items 1 to 98 of
98
with 100 items per page.
- W3170842411 abstract "Vision language navigation is the task that requires an agent to navigate through a 3D environment based on natural language instructions. One key challenge in this task is to ground instructions with the current visual information that the agent perceives. Most of the existing work employs soft attention over individual words to locate the instruction required for the next action. However, different words have different functions in a sentence (e.g., modifiers convey attributes, verbs convey actions). Syntax information like dependencies and phrase structures can aid the agent to locate important parts of the instruction. Hence, in this paper, we propose a navigation agent that utilizes syntax information derived from a dependency tree to enhance alignment between the instruction and the current visual scenes. Empirically, our agent outperforms the baseline model that does not use syntax information on the Room-to-Room dataset, especially in the unseen environment. Besides, our agent achieves the new state-of-the-art on Room-Across-Room dataset, which contains instructions in 3 languages (English, Hindi, and Telugu). We also show that our agent is better at aligning instructions with the current visual information via qualitative visualizations." @default.
- W3170842411 created "2021-06-22" @default.
- W3170842411 creator A5001987532 @default.
- W3170842411 creator A5074267258 @default.
- W3170842411 creator A5087164997 @default.
- W3170842411 date "2021-01-01" @default.
- W3170842411 modified "2023-09-25" @default.
- W3170842411 title "Improving Cross-Modal Alignment in Vision Language Navigation via Syntactic Information" @default.
- W3170842411 cites W2117539524 @default.
- W3170842411 cites W2123442489 @default.
- W3170842411 cites W2194775991 @default.
- W3170842411 cites W2608787653 @default.
- W3170842411 cites W2884565639 @default.
- W3170842411 cites W2890902815 @default.
- W3170842411 cites W2926977875 @default.
- W3170842411 cites W2948008321 @default.
- W3170842411 cites W2951973805 @default.
- W3170842411 cites W2958574008 @default.
- W3170842411 cites W2962744691 @default.
- W3170842411 cites W2963355447 @default.
- W3170842411 cites W2963575117 @default.
- W3170842411 cites W2963580443 @default.
- W3170842411 cites W2963661253 @default.
- W3170842411 cites W2963726321 @default.
- W3170842411 cites W2963846044 @default.
- W3170842411 cites W2964043796 @default.
- W3170842411 cites W2964167098 @default.
- W3170842411 cites W2964343989 @default.
- W3170842411 cites W2964935470 @default.
- W3170842411 cites W2970340522 @default.
- W3170842411 cites W2972795848 @default.
- W3170842411 cites W2979727876 @default.
- W3170842411 cites W2981799368 @default.
- W3170842411 cites W2987914945 @default.
- W3170842411 cites W2991186560 @default.
- W3170842411 cites W2995538426 @default.
- W3170842411 cites W3010436080 @default.
- W3170842411 cites W3016321512 @default.
- W3170842411 cites W3023306062 @default.
- W3170842411 cites W3029418112 @default.
- W3170842411 cites W3034253961 @default.
- W3170842411 cites W3034376488 @default.
- W3170842411 cites W3037109418 @default.
- W3170842411 cites W3100923070 @default.
- W3170842411 cites W3105521436 @default.
- W3170842411 cites W3106129629 @default.
- W3170842411 cites W3106641651 @default.
- W3170842411 cites W3109380382 @default.
- W3170842411 doi "https://doi.org/10.18653/v1/2021.naacl-main.82" @default.
- W3170842411 hasPublicationYear "2021" @default.
- W3170842411 type Work @default.
- W3170842411 sameAs 3170842411 @default.
- W3170842411 citedByCount "6" @default.
- W3170842411 countsByYear W31708424112022 @default.
- W3170842411 countsByYear W31708424112023 @default.
- W3170842411 crossrefType "proceedings-article" @default.
- W3170842411 hasAuthorship W3170842411A5001987532 @default.
- W3170842411 hasAuthorship W3170842411A5074267258 @default.
- W3170842411 hasAuthorship W3170842411A5087164997 @default.
- W3170842411 hasBestOaLocation W31708424111 @default.
- W3170842411 hasConcept C107457646 @default.
- W3170842411 hasConcept C154945302 @default.
- W3170842411 hasConcept C162324750 @default.
- W3170842411 hasConcept C187736073 @default.
- W3170842411 hasConcept C195324797 @default.
- W3170842411 hasConcept C204321447 @default.
- W3170842411 hasConcept C2777530160 @default.
- W3170842411 hasConcept C2780451532 @default.
- W3170842411 hasConcept C41008148 @default.
- W3170842411 hasConcept C60048249 @default.
- W3170842411 hasConceptScore W3170842411C107457646 @default.
- W3170842411 hasConceptScore W3170842411C154945302 @default.
- W3170842411 hasConceptScore W3170842411C162324750 @default.
- W3170842411 hasConceptScore W3170842411C187736073 @default.
- W3170842411 hasConceptScore W3170842411C195324797 @default.
- W3170842411 hasConceptScore W3170842411C204321447 @default.
- W3170842411 hasConceptScore W3170842411C2777530160 @default.
- W3170842411 hasConceptScore W3170842411C2780451532 @default.
- W3170842411 hasConceptScore W3170842411C41008148 @default.
- W3170842411 hasConceptScore W3170842411C60048249 @default.
- W3170842411 hasLocation W31708424111 @default.
- W3170842411 hasLocation W31708424112 @default.
- W3170842411 hasOpenAccess W3170842411 @default.
- W3170842411 hasPrimaryLocation W31708424111 @default.
- W3170842411 hasRelatedWork W1573537589 @default.
- W3170842411 hasRelatedWork W159132833 @default.
- W3170842411 hasRelatedWork W2033261979 @default.
- W3170842411 hasRelatedWork W2293457016 @default.
- W3170842411 hasRelatedWork W2464738873 @default.
- W3170842411 hasRelatedWork W2567044968 @default.
- W3170842411 hasRelatedWork W2977842567 @default.
- W3170842411 hasRelatedWork W3082447286 @default.
- W3170842411 hasRelatedWork W87581401 @default.
- W3170842411 hasRelatedWork W1872130062 @default.
- W3170842411 isParatext "false" @default.
- W3170842411 isRetracted "false" @default.
- W3170842411 magId "3170842411" @default.
- W3170842411 workType "article" @default.