---
_id: '34083'
abstract:
- lang: eng
text: In the context of language learning, feedback comment generation is the task
of generating hints or explanatory notes for learner texts that help understand
why a part of text is erroneous. This paper presents our approach to the Feedback
Comment Generation Shared Task, collocated with the 16th International Natural
Language Generation Conference (INLG 2023). The approach augments the generation
of feedback comments by a self-supervised identification of feedback types in
a multitask-learning setting. Within the shared task, other approaches performed
more effective, yet the combined modeling of feedback type classification and
feedback comment generation is superior to performing feedback generation only.
author:
- first_name: Maja
full_name: Stahl, Maja
id: '77647'
last_name: Stahl
- first_name: Henning
full_name: Wachsmuth, Henning
id: '3900'
last_name: Wachsmuth
citation:
ama: 'Stahl M, Wachsmuth H. Identifying Feedback Types to Augment Feedback Comment
Generation. In: Proceedings of the 16th International Natural Language Generation
Conference.'
apa: Stahl, M., & Wachsmuth, H. (n.d.). Identifying Feedback Types to Augment
Feedback Comment Generation. Proceedings of the 16th International Natural
Language Generation Conference. 16th International Natural Language Generation
Conference.
bibtex: '@inproceedings{Stahl_Wachsmuth, title={Identifying Feedback Types to Augment
Feedback Comment Generation}, booktitle={Proceedings of the 16th International
Natural Language Generation Conference}, author={Stahl, Maja and Wachsmuth, Henning}
}'
chicago: Stahl, Maja, and Henning Wachsmuth. “Identifying Feedback Types to Augment
Feedback Comment Generation.” In Proceedings of the 16th International Natural
Language Generation Conference, n.d.
ieee: M. Stahl and H. Wachsmuth, “Identifying Feedback Types to Augment Feedback
Comment Generation,” presented at the 16th International Natural Language Generation
Conference.
mla: Stahl, Maja, and Henning Wachsmuth. “Identifying Feedback Types to Augment
Feedback Comment Generation.” Proceedings of the 16th International Natural
Language Generation Conference.
short: 'M. Stahl, H. Wachsmuth, in: Proceedings of the 16th International Natural
Language Generation Conference, n.d.'
conference:
name: 16th International Natural Language Generation Conference
date_created: 2022-11-15T08:47:57Z
date_updated: 2022-11-15T08:48:01Z
extern: '1'
language:
- iso: eng
publication: Proceedings of the 16th International Natural Language Generation Conference
publication_status: accepted
status: public
title: Identifying Feedback Types to Augment Feedback Comment Generation
type: conference
user_id: '77647'
year: '2023'
...
---
_id: '34051'
abstract:
- lang: eng
text: An argument is a constellation of premises reasoning towards a certain conclusion.
The automatic generation of conclusions is becoming a very prominent task, raising
the need for automatic measures to assess the quality of these generated conclusions.
The SharedTask at the 9th Workshop on Argument Mining proposes a new task to assess
the novelty and validity of a conclusion given a set of premises. In this paper,
we present a multitask learning approach that transfers the knowledge learned
from the natural language inference task to the tasks at hand. Evaluation results
indicate the importance of both knowledge transfer and joint learning, placing
our approach in the fifth place with strong results compared to baselines.
author:
- first_name: Milad
full_name: Alshomary, Milad
id: '73059'
last_name: Alshomary
- first_name: Maja
full_name: Stahl, Maja
id: '77647'
last_name: Stahl
citation:
ama: 'Alshomary M, Stahl M. Argument Novelty and Validity Assessment via Multitask
and Transfer Learning. In: Proceedings of the 9th Workshop on Argument Mining.
International Conference on Computational Linguistics; 2022:111–114.'
apa: Alshomary, M., & Stahl, M. (2022). Argument Novelty and Validity Assessment
via Multitask and Transfer Learning. Proceedings of the 9th Workshop on Argument
Mining, 111–114.
bibtex: '@inproceedings{Alshomary_Stahl_2022, place={Online and in Gyeongju, Republic
of Korea}, title={Argument Novelty and Validity Assessment via Multitask and Transfer
Learning}, booktitle={Proceedings of the 9th Workshop on Argument Mining}, publisher={International
Conference on Computational Linguistics}, author={Alshomary, Milad and Stahl,
Maja}, year={2022}, pages={111–114} }'
chicago: 'Alshomary, Milad, and Maja Stahl. “Argument Novelty and Validity Assessment
via Multitask and Transfer Learning.” In Proceedings of the 9th Workshop on
Argument Mining, 111–114. Online and in Gyeongju, Republic of Korea: International
Conference on Computational Linguistics, 2022.'
ieee: M. Alshomary and M. Stahl, “Argument Novelty and Validity Assessment via Multitask
and Transfer Learning,” in Proceedings of the 9th Workshop on Argument Mining,
2022, pp. 111–114.
mla: Alshomary, Milad, and Maja Stahl. “Argument Novelty and Validity Assessment
via Multitask and Transfer Learning.” Proceedings of the 9th Workshop on Argument
Mining, International Conference on Computational Linguistics, 2022, pp. 111–114.
short: 'M. Alshomary, M. Stahl, in: Proceedings of the 9th Workshop on Argument
Mining, International Conference on Computational Linguistics, Online and in Gyeongju,
Republic of Korea, 2022, pp. 111–114.'
date_created: 2022-11-10T09:51:45Z
date_updated: 2022-11-15T08:49:10Z
language:
- iso: eng
page: 111–114
place: Online and in Gyeongju, Republic of Korea
publication: Proceedings of the 9th Workshop on Argument Mining
publication_status: published
publisher: International Conference on Computational Linguistics
status: public
title: Argument Novelty and Validity Assessment via Multitask and Transfer Learning
type: conference
user_id: '77647'
year: '2022'
...
---
_id: '34082'
abstract:
- lang: eng
text: Gender bias may emerge from an unequal representation of agency and power,
for example, by portraying women frequently as passive and powerless ("She accepted
her future'') and men as proactive and powerful ("He chose his future''). When
language models learn from respective texts, they may reproduce or even amplify
the bias. An effective way to mitigate bias is to generate counterfactual sentences
with opposite agency and power to the training. Recent work targeted agency-specific
verbs from a lexicon to this end. We argue that this is insufficient, due to the
interaction of agency and power and their dependence on context. In this paper,
we thus develop a new rewriting model that identifies verbs with the desired agency
and power in the context of the given sentence. The verbs' probability is then
boosted to encourage the model to rewrite both connotations jointly. According
to automatic metrics, our model effectively controls for power while being competitive
in agency to the state of the art. In our evaluation, human annotators favored
its counterfactuals in terms of both connotations, also deeming its meaning preservation
better.
author:
- first_name: Maja
full_name: Stahl, Maja
id: '77647'
last_name: Stahl
- first_name: Maximilian
full_name: Spliethöver, Maximilian
id: '84035'
last_name: Spliethöver
orcid: 0000-0003-4364-1409
- first_name: Henning
full_name: Wachsmuth, Henning
id: '3900'
last_name: Wachsmuth
citation:
ama: 'Stahl M, Spliethöver M, Wachsmuth H. To Prefer or to Choose? Generating Agency
and Power Counterfactuals Jointly for Gender Bias Mitigation. In: Proceedings
of the Fifth Workshop on Natural Language Processing and Computational Social
Science.'
apa: Stahl, M., Spliethöver, M., & Wachsmuth, H. (n.d.). To Prefer or to Choose?
Generating Agency and Power Counterfactuals Jointly for Gender Bias Mitigation.
Proceedings of the Fifth Workshop on Natural Language Processing and Computational
Social Science. Fifth Workshop on NLP and Computational Social Science (NLP+CSS)
At EMNLP 2022, Abu Dhabi, United Arab Emirates.
bibtex: '@inproceedings{Stahl_Spliethöver_Wachsmuth, place={Abu Dhabi, United Arab
Emirates}, title={To Prefer or to Choose? Generating Agency and Power Counterfactuals
Jointly for Gender Bias Mitigation}, booktitle={Proceedings of the Fifth Workshop
on Natural Language Processing and Computational Social Science}, author={Stahl,
Maja and Spliethöver, Maximilian and Wachsmuth, Henning} }'
chicago: Stahl, Maja, Maximilian Spliethöver, and Henning Wachsmuth. “To Prefer
or to Choose? Generating Agency and Power Counterfactuals Jointly for Gender Bias
Mitigation.” In Proceedings of the Fifth Workshop on Natural Language Processing
and Computational Social Science. Abu Dhabi, United Arab Emirates, n.d.
ieee: M. Stahl, M. Spliethöver, and H. Wachsmuth, “To Prefer or to Choose? Generating
Agency and Power Counterfactuals Jointly for Gender Bias Mitigation,” presented
at the Fifth Workshop on NLP and Computational Social Science (NLP+CSS) At EMNLP
2022, Abu Dhabi, United Arab Emirates.
mla: Stahl, Maja, et al. “To Prefer or to Choose? Generating Agency and Power Counterfactuals
Jointly for Gender Bias Mitigation.” Proceedings of the Fifth Workshop on Natural
Language Processing and Computational Social Science.
short: 'M. Stahl, M. Spliethöver, H. Wachsmuth, in: Proceedings of the Fifth Workshop
on Natural Language Processing and Computational Social Science, Abu Dhabi, United
Arab Emirates, n.d.'
conference:
location: Abu Dhabi, United Arab Emirates
name: Fifth Workshop on NLP and Computational Social Science (NLP+CSS) At EMNLP
2022
date_created: 2022-11-15T08:29:26Z
date_updated: 2022-11-18T08:22:56Z
extern: '1'
language:
- iso: eng
place: Abu Dhabi, United Arab Emirates
publication: Proceedings of the Fifth Workshop on Natural Language Processing and
Computational Social Science
publication_status: accepted
quality_controlled: '1'
status: public
title: To Prefer or to Choose? Generating Agency and Power Counterfactuals Jointly
for Gender Bias Mitigation
type: conference
user_id: '77647'
year: '2022'
...