publications

2026

1. EMNLP

   Who Flips? Self- and Cross-Model Counterarguments Reveal Answer Instability in LLMs

   Nafiseh Nikeghbal, Amir Hossein Kargaran, Shaghayegh Kolli, and 1 more author

   In Findings of the Association for Computational Linguistics: EMNLP 2026, Nov 2026

   Abs Bib

   We study how robust large language models are to counterarguments, showing that self- and cross-model counterarguments can flip model answers and reveal underlying answer instability.

   @inproceedings{nikeghbal2026whoflips,
     title = {Who Flips? Self- and Cross-Model Counterarguments Reveal Answer Instability in LLMs},
     author = {Nikeghbal, Nafiseh and Kargaran, Amir Hossein and Kolli, Shaghayegh and Diesner, Jana},
     booktitle = {Findings of the Association for Computational Linguistics: EMNLP 2026},
     year = {2026},
     month = nov,
   }

2. ICML-W

   StylisticBias: A Few Human Visual Cues Drive Most Social Bias in MLLMs

   Shaghayegh Kolli, Timo Cavelius, Nafiseh Nikeghbal, and 2 more authors

   In AI for Good Workshop (AI4GOOD) at ICML 2026, Jul 2026

   Abs Bib

   We show that a small set of human visual cues accounts for most of the social bias exhibited by multimodal large language models (MLLMs).

   @inproceedings{kolli2026stylisticbias,
     title = {StylisticBias: A Few Human Visual Cues Drive Most Social Bias in MLLMs},
     author = {Kolli, Shaghayegh and Cavelius, Timo and Nikeghbal, Nafiseh and Dalal, Samantha and Diesner, Jana},
     booktitle = {AI for Good Workshop (AI4GOOD) at ICML 2026},
     year = {2026},
     month = jul
   }

3. Preprint

   Neuron-Level Interventions for Gendered and Gender-Neutral Generation in Language Models

   Zhuoran You, Nafiseh Nikeghbal, and Jana Diesner

   Under review, 2026

   Abs Bib

   We identify and intervene on neurons responsible for gendered generation in language models, enabling controllable gendered and gender-neutral text generation.

   @article{you2026neuron,
     title = {Neuron-Level Interventions for Gendered and Gender-Neutral Generation in Language Models},
     author = {You, Zhuoran and Nikeghbal, Nafiseh and Diesner, Jana},
     journal = {Under review},
     year = {2026}
   }

4. Preprint

   GlotOCR Bench: OCR Models Still Struggle Beyond a Handful of Unicode Scripts

   Amir Hossein Kargaran, Nafiseh Nikeghbal, Jana Diesner, and 2 more authors

   Under review, 2026

   Abs Bib

   We introduce GlotOCR Bench, a multilingual OCR benchmark showing that current OCR models still struggle far beyond a small set of high-resource Unicode scripts.

   @article{kargaran2026glotocr,
     title = {GlotOCR Bench: OCR Models Still Struggle Beyond a Handful of Unicode Scripts},
     author = {Kargaran, Amir Hossein and Nikeghbal, Nafiseh and Diesner, Jana and Yvon, Fran{\c{c}}ois and Sch{\"u}tze, Hinrich},
     journal = {Under review},
     year = {2026}
   }

2025

1. Preprint

   Insights from the ICLR Peer Review and Rebuttal Process

   Amir Hossein Kargaran, Nafiseh Nikeghbal, Jiaxin Yang, and 1 more author

   arXiv preprint, 2025

   Abs Bib

   We analyze large-scale data from the ICLR peer review and rebuttal process to surface trends in scores, reviewer behavior, and the effect of rebuttals on outcomes.

   @article{kargaran2025iclr,
     title = {Insights from the ICLR Peer Review and Rebuttal Process},
     author = {Kargaran, Amir Hossein and Nikeghbal, Nafiseh and Yang, Jiaxin and Ousidhoum, Nedjma},
     journal = {arXiv preprint},
     year = {2025}
   }

2. ACL

   MEXA: Multilingual Evaluation of English-Centric LLMs via Cross-Lingual Alignment

   Amir Hossein Kargaran, Ali Modarressi, Nafiseh Nikeghbal, and 3 more authors

   In Findings of the Association for Computational Linguistics: ACL 2025, Jul 2025

   Abs arXiv Bib HTML Code

   MEXA assesses the multilingual capabilities of English-centric LLMs using parallel sentences, computing the alignment between English and non-English languages in intermediate layers to estimate performance across many languages.

   @inproceedings{kargaran2025mexa,
     title = {MEXA: Multilingual Evaluation of English-Centric LLMs via Cross-Lingual Alignment},
     author = {Kargaran, Amir Hossein and Modarressi, Ali and Nikeghbal, Nafiseh and Diesner, Jana and Yvon, Fran{\c{c}}ois and Sch{\"u}tze, Hinrich},
     booktitle = {Findings of the Association for Computational Linguistics: ACL 2025},
     pages = {27001--27023},
     year = {2025},
     month = jul,
     publisher = {Association for Computational Linguistics},
     address = {Vienna, Austria},
   }

3. EMNLP

   CoBia: Constructed Conversations Can Trigger Otherwise Concealed Societal Biases in LLMs

   Nafiseh Nikeghbal, Amir Hossein Kargaran, and Jana Diesner

   In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP), Nov 2025

   Abs arXiv Bib HTML Code

   CoBia is a suite of lightweight adversarial attacks that construct conversations in which a model utters a biased claim about a social group, then tests whether the model can recover and reject biased follow-ups. Across 11 open and proprietary LLMs and six socio-demographic categories, constructed conversations reliably reveal biases that standard safety checks miss.

   @inproceedings{nikeghbal2025cobia,
     title = {CoBia: Constructed Conversations Can Trigger Otherwise Concealed Societal Biases in LLMs},
     author = {Nikeghbal, Nafiseh and Kargaran, Amir Hossein and Diesner, Jana},
     booktitle = {Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
     year = {2025},
     month = nov,
     publisher = {Association for Computational Linguistics},
   }

2024

1. MSR

   GIRT-Model: Automated Generation of Issue Report Templates

   Nafiseh Nikeghbal, Amir Hossein Kargaran, and Abbas Heydarnoori

   In Proceedings of the 21st IEEE/ACM International Conference on Mining Software Repositories (MSR), 2024

   Abs Bib Code

   GIRT-Model is an assistant that automatically generates structured Issue Report Templates (IRTs) for GitHub repositories based on developer instructions.

   @inproceedings{nikeghbal2024girtmodel,
     title = {GIRT-Model: Automated Generation of Issue Report Templates},
     author = {Nikeghbal, Nafiseh and Kargaran, Amir Hossein and Heydarnoori, Abbas},
     booktitle = {Proceedings of the 21st IEEE/ACM International Conference on Mining Software Repositories (MSR)},
     year = {2024},
     publisher = {IEEE/ACM},
   }

2023

1. MSR

   GIRT-Data: Sampling GitHub Issue Report Templates

   Nafiseh Nikeghbal, Amir Hossein Kargaran, Abbas Heydarnoori, and 1 more author

   In Proceedings of the 20th IEEE/ACM International Conference on Mining Software Repositories (MSR), 2023

   Abs Bib Code

   GIRT-Data is the first and largest dataset of Issue Report Templates (IRTs) collected from GitHub, released together with tooling to study and support structured issue reporting.

   @inproceedings{nikeghbal2023girtdata,
     title = {GIRT-Data: Sampling GitHub Issue Report Templates},
     author = {Nikeghbal, Nafiseh and Kargaran, Amir Hossein and Heydarnoori, Abbas and Sch{\"u}tze, Hinrich},
     booktitle = {Proceedings of the 20th IEEE/ACM International Conference on Mining Software Repositories (MSR)},
     year = {2023},
     publisher = {IEEE/ACM},
   }

2. Preprint

   MenuCraft: Interactive Menu System Design with Large Language Models

   Amir Hossein Kargaran, Nafiseh Nikeghbal, Abbas Heydarnoori, and 1 more author

   arXiv preprint, 2023

   Abs Bib

   MenuCraft is an interactive tool for menu system design that leverages large language models to collaboratively design and refine user-interface menus through dialogue.

   @article{kargaran2023menucraft,
     title = {MenuCraft: Interactive Menu System Design with Large Language Models},
     author = {Kargaran, Amir Hossein and Nikeghbal, Nafiseh and Heydarnoori, Abbas and Sch{\"u}tze, Hinrich},
     journal = {arXiv preprint},
     year = {2023}
   }