Skip to content

AMR extension for the spatial domain, with grounded frame of reference tracking

License

Notifications You must be signed in to change notification settings

cu-clear/Spatial-AMR

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 

Repository files navigation

Spatial-AMR

Spatial AMR extends the 2019 version of Abstract Meaning Representation (AMR) to allow fine-grained spatial semantics and pragmatics from grounded, multimodal corpora to be incorporated into the graphs; there is a special focus on annotating linguistic frame of reference (FoR).

Spatial AMR was developed using the Minecraft Dialogue Corpus, in which sets of two particpants-- a Builder and and Architect-- collaborate in building a wide array of block structures in Minecraft. In recording the dialogues, the MALMO platform recorded entity locations and orientations in the 3D, Cartesian framework of the Minecraft environment, as well as screenshots from 6 angles for each dialogue act and each block placement action. This grounding data is represented in the annotations, but is also available in raw form in the Minecraft Dialogue Corpus links below.

Resources:

Spatial AMR paper:

https://par.nsf.gov/biblio/10179909 Bonn, Julia, Martha Palmer, Jon Cai, and Kristin Wright-Bettner. "Spatial AMR: Expanded spatial annotation in the context of a grounded Minecraft corpus." In Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020),. 2020.

Rolesets:

An interface for viewing the English PropBank Lexicon of Rolesets:

http://ska.tjemye.rs/propbank/development-frames/

  • In the upper-left drop-down menu, select 'AMR Spatial 1.0'. This will filter the rolesets to the full set of English rolesets used in Spatial AMR annotation, including those with special spatial semantics.
  • A limited inventory of just the spatial-semantics rolesets will presented separately in the Spatial AMR Github repository-- coming soon

The Minecraft Dialogue Corpus:

Anjali Narayan-Chen, Prashant Jayannavar, and Julia Hockenmaier. 2019. Collaborative Dialogue in Minecraft. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 5405–5415, Florence, Italy. Association for Computational Linguistics. https://aclanthology.org/P19-1537/

The dialogue data

https://github.com/prashant-jayan21/minecraft-dialogue

Related Publications:

Kokel, Harsha, M. Das, Rakibul Islam, Julia Bonn, Jon Z. Cai, Soham Dan, Anjali Narayan-Chen, Prashant Jayannavar, Janardhan Rao Doppa, J. Hockenmaier, Sriraam Natarajan, Martha Palmer and Dan Roth. “Lara – Human-guided collaborative problem solver: Effective integration of learning, reasoning and communication.” (2022).

Dan, Soham, Parisa Kordjamshidi, Julia Bonn, Archna Bhatia, Zheng Jon Cai, Martha Palmer and Dan Roth. “From Spatial Relations to Spatial Configurations.” International Conference on Language Resources and Evaluation (2020).

Releases

No releases published

Packages

No packages published