Postdoctoral Researcher · University of Bonn

Akbar Karimi

CAISA Lab — Conversational AI and Social Analytics

Email Scholar ORCID GitHub
Akbar Karimi

My research sits at the intersection of robustness and language: I study how large language models break under adversarial pressure — arithmetic attacks, distributional shift, multi-hop reasoning failures — and build methods to make them more resilient.

Before this, I completed my PhD at the University of Parma under Prof. Andrea Prati at IMP Lab, where I developed adversarial learning and data augmentation techniques for more robust language models.

News

Dec 2025 Attending Eurips Conference and presenting ArithmAttack and Multi-hop Reasoning with Hyperbolic Representations at the Ellis UnConference.
Nov 2025 New preprint: More Agents Helps but Adversarial Robustness Gap Persists.
Sep 2025 Attending the ECMLPKDD Conference in Porto to present the results from the Colliding with Adversaries Challenge.
Aug 2025 Attending Interspeech Conference in Rotterdam to present our recent work on speech data augmentation for German dialects.
Aug 2025 Attending ACL Conference in Vienna to present ArithmAttack and Multi-hop Reasoning with Hyperbolic Representations.

Publications

2026
More Agents Improve Math Problem Solving but Adversarial Robustness Gap Persists
Findings of ACL 2026
PDF
2026
MiniFool: Physics-Constraint-Aware Minimizer-Based Adversarial Attacks in Deep Neural Networks
The European Physical Journal C
PDF
2026
Speaker Normalization via Voice Conversion Reveals a Human–Machine Dissociation in Dialect Classification
DialRes Workshop @ LREC 2026
PDF
2026
Can LLM Agents Identify Spoken Dialects like a Linguist?
DialRes Workshop @ LREC 2026
PDF
2026
Label-Consistent Data Generation for Aspect-Based Sentiment Analysis Using LLM Agents
WASSA Workshop @ EACL 2026
PDF
2026
Shapes are not enough: CONSERVAttack and its use for finding vulnerabilities and uncertainties in machine learning applications
arXiv preprint
PDF
2025
Encoder Fine-tuning with Stochastic Sampling Outperforms Open-weight GPT in Astronomy Knowledge Extraction
WASP Workshop @ IJCNLP-AACL 2025
PDF
2025
Enforcing Fundamental Relations via Adversarial Attacks on Input Parameter Correlations
Computing and Software for Big Science
PDF
2025
Colliding with Adversaries: A Challenge on Robust Learning in High Energy Physics at ECML PKDD 2025
ECML PKDD 2025
PDF
2025
EDAudio: Easy Data Augmentation for Dialectal Audio
RANLP 2025
PDF
2025
Improving Low-Resource Dialect Classification Using Retrieval-based Voice Conversion
Interspeech 2025
PDF
2025
Multi-Hop Reasoning for Question Answering with Hyperbolic Representations
Findings of ACL
PDF
2025
ArithmAttack: Evaluating Robustness of LLMs to Noisy Context in Math Problem Solving
LLMSEC Workshop @ ACL
PDF
2025
Explainable Hallucination through Natural Language Inference Mapping
Findings of ACL
PDF
2024
Exploring Robustness of Multilingual LLMs on Real-World Noisy Data
Eighth Widening NLP Workshop (WiNLP 2024) Phase II
PDF
2024
Do Multilingual Large Language Models Mitigate Stereotype Bias?
C3NLP Workshop @ ACL
PDF
2023
CAISA at SemEval-2023 Task 8: Counterfactual Data Augmentation for Mitigating Class Imbalance in Causal Claim Identification
SemEval 2023
PDF
2022
CAISA@SMM4H’22: Robust Cross-Lingual Detection of Disease Mentions on Social Media with Adversarial Methods
SMM4H 2022
PDF
2022
Self-Balanced R-CNN for Instance Segmentation
JVCIR
PDF
2022
Aspect-Based Emotion Analysis and Multimodal Coreference: A Case Study of Customer Comments on Adidas Instagram Posts
LREC 2022
PDF
2022
Improving Localization for Semi-Supervised Object Detection
ICIAP 2022
PDF
2021
Improving BERT Performance for Aspect-Based Sentiment Analysis
ICNLSP 2021
PDF
2021
Recursively Refined R-CNN: Instance Segmentation with Self-RoI Rebalancing
CAIP 2021
PDF
2021
AEDA: An Easier Data Augmentation Technique for Text Classification
Findings of EMNLP
PDF
2021
Uniparma at SemEval-2021 Task 5: Toxic Spans Detection Using CharacterBERT and Bag-of-Words Model
SemEval 2021
PDF
2020
A Novel Region of Interest Extraction Layer for Instance Segmentation
ICPR 2020
PDF
2020
Adversarial Training for Aspect-Based Sentiment Analysis with BERT
ICPR
PDF
2018
Extracting an English-Persian Parallel Corpus from Comparable Corpora
LREC 2018
PDF

Teaching

2025
Dialog Systems
Undergraduate course · University of Bonn
2024
Dialog Systems
Undergraduate course · University of Bonn
2023
Introduction to Natural Language Processing
Undergraduate course · University of Marburg
2022
Dialog Systems
Undergraduate course · University of Marburg