Eight new publications by the Heidelberg NLP group, collaborating researchers and CL students:
We are delighted to announce four papers on interpretation and reasoning in NLU and NLG using commonsense knowledge, to appear at ACL-IJCNLP 2021, at *SEM 2021, at the DeeLIO Workshop @NAACL 2021 and in IWCS 2021, the International Conference on Computational Semantics in Groningen.
- Debjit Paul and Anette Frank: COINS: Learning to Generate COntextualized Inference Rules for Narrative Story Completion. To appear at ACL-IJCNLP 2021. arXiv preprint
- Maria Becker, Siting Liang and Anette Frank: Reconstructing Implicit Knowledge with Language Models. To appear at: Deep Learning Inside Out (DeeLIO): The Workshop on Knowledge Extraction and Integration for Deep Learning Architectures @ NAACL 2021. Earning the Best Paper Award
- Maria Becker, Katharina Korfhage, Debjit Paul and Anette Frank: CO-NNECT: A Framework for Revealing Commonsense Knowledge Paths as Explicitations of Implicit Knowledge in Texts. To appear in IWCS 2021. arXiv preprint
- Debjit Paul and Anette Frank: Generating Hypothetical Events for Abductive Inference. To appear in: The Tenth Joint Conference on Lexical and Computational Semantics, *SEM 2021, collocated with ACL-IJCNLP 2021.
There is also new research on multilingual AMR, arising from a seminar in our curriculum:
- Sarah Uhrig, Yoalli Garcia, Juri Opitz and Anette Frank: Translate, then Parse! A Strong Baseline for Cross-Lingual AMR Parsing. To appear in: IWPT 2021: The 17th International Conference on Parsing Technologies @ ACL-IJCNLP 2021.
And three more papers all related to Multimodal Vision and Language Processing will appear in the Workshop on Advances in Language and Vision Research (ALVR) @NAACL 2021 and in the Multimodal Semantic Representations Workshop @IWCS 2021.
- Julia Suter, Letitia Parcalabescu and Anette Frank: Grounding Plural Phrases: Countering Evaluation Biases by Individuation. To appear in ALVR @NAACL 2021.
- Letitia Parcalabescu, Nils Trost and Anette Frank: What is Multimodality? First Workshop on Multimodal Semantic Representations. To appear in MMSR @IWCS and to be presented in ALVR@NAACL 2021. arXiv preprint
- Letitia Parcalabescu, Albert Gatt, Anette Frank and Iacer Calixto: Seeing Past Words: Testing the Cross-Modal Capabilities of pretrained V&L models. To appear in MMSR @IWCS 2021 and also being presented in Repl4NLP @ ACL-IJCNLP 2021. arXiv preprint
Congratulations to all authors!