Amirhossein Dabiriaghdam
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Amirhossein Dabiriaghdam has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Mohammadreza Bakhtiari
University of British Columbia Stony Brook University Yadollah Yaghoobzadeh University of Tehran
2 shared publications Shayan Vassef
University of British Columbia Stony Brook University Yadollah Yaghoobzadeh University of Tehran
2 shared publications- Yadollah Yaghoobzadeh
School of Electrical and Computer Engineering University of Tehran Tehran Iran
2 shared publications - Lele Wang
Department of Electrical and Computer Engineering, The University of British Columbia, Vancouver, BC, Canada
1 shared publication
Latest Publications
- Open Access
SOI Matters: Analyzing Multi-Setting Training Dynamics in Pretrained Language Models via Subsets of Interest
This work investigates the impact of multi-task, multi-lingual, and multi-source learning approaches on the robustness and performance of pretrained language models. To enhance this analysis, we introduce Subsets of Interest (SOI), a novel categorization framework that identifies six distinct learning behavior patterns during training, including forgettable examples, unlearned examples, and always correct examples. Through SOI transition heatmaps and dataset cartography visualization, we analyze how examples shift between these categories when transitioning from single-setting to multi-setting configurations. We perform comprehensive experiments across three parallel comparisons: multi-task vs. single-task learning using English tasks (entailment, paraphrase, sentiment), multi-source vs. single-source learning using sentiment analysis datasets, and multi-lingual vs. single-lingual learning using intent classification in French, English, and Persian. Our results demonstrate that multi-source learning consistently improves out-of-distribution performance by up to 7%, while multi-task learning shows mixed results with notable gains in similar task combinations. We further introduce a two-stage fine-tuning approach where the second stage leverages SOI-based subset selection to achieve additional performance improvements. These findings provide new insights into training dynamics and offer practical approaches for optimizing multi-setting language model performance.
Open AccessSOI Matters: Analyzing Multi-Setting Training Dynamics in Pretrained Language Models via Subsets of Interest
This work investigates the impact of multi-task, multi-lingual, and multi-source learning approaches on the robustness and performance of pretrained language models. To enhance this analysis, we introduce Subsets of Interest (SOI), a novel categorization framework that identifies six distinct learning behavior patterns during training, including forgettable examples, unlearned examples, and always correct examples. Through SOI transition heatmaps and dataset cartography visualization, we analyze how examples shift between these categories when transitioning from single-setting to multi-setting configurations. We perform comprehensive experiments across three parallel comparisons: multi-task vs. single-task learning using English tasks (entailment, paraphrase, sentiment), multi-source vs. single-source learning using sentiment analysis datasets, and multi-lingual vs. single-lingual learning using intent classification in French, English, and Persian. Our results demonstrate that multi-source learning consistently improves out-of-distribution performance by up to 7%, while multi-task learning shows mixed results with notable gains in similar task combinations. We further introduce a two-stage fine-tuning approach where the second stage leverages SOI-based subset selection to achieve additional performance improvements. These findings provide new insights into training dynamics and offer practical approaches for optimizing multi-setting language model performance.
- Open Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 322-335
SOI Matters: Analyzing Multi-Setting Training Dynamics in Pretrained Language Models via Subsets of Interest
Shayan Vassef, Amirhossein Dabiriaghdam, Mohammadreza Bakhtiari, Yadollah Yaghoobzadeh. Proceedings of the 5th Workshop on Multilingual Representation Learning (MRL 2025). 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 322-335
Open AccessSOI Matters: Analyzing Multi-Setting Training Dynamics in Pretrained Language Models via Subsets of Interest
Shayan Vassef, Amirhossein Dabiriaghdam, Mohammadreza Bakhtiari, Yadollah Yaghoobzadeh. Proceedings of the 5th Workshop on Multilingual Representation Learning (MRL 2025). 2025.
- Open Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 30773-30794
SimMark: A Robust Sentence-Level Similarity-Based Watermarking Algorithm for Large Language Models
Amirhossein Dabiriaghdam, Lele Wang. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 30773-30794
Open AccessSimMark: A Robust Sentence-Level Similarity-Based Watermarking Algorithm for Large Language Models
Amirhossein Dabiriaghdam, Lele Wang. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.