Papers by Gözde Gül Şahin
LINSPECTOR WEB: A Multilingual Probing Suite for Word Representations (D19-3)
Copied to clipboard
| Challenge: | LINSPECTOR WEB is an open source multilingual inspector to analyze word embeddings. |
| Approach: | They propose to use LINSPECTOR WEB to analyze word embeddings in 28 languages. |
| Outcome: | The system performs 16 simple linguistic probing tasks for a diverse set of 28 languages. |
Data Augmentation via Dependency Tree Morphing for Low-Resource Languages (D18-1)
Copied to clipboard
| Challenge: | Lack of sizable training datasets leads to poor performance in low-resource languages. |
| Approach: | They propose two techniques to augment training sets of low-resource languages using dependency trees. |
| Outcome: | The proposed methods improve on the training datasets for low-resource languages. |
PARADISE: Evaluating Implicit Planning Skills of Language Models with Procedural Warnings and Tips Dataset (2024.findings-acl)
Copied to clipboard
| Challenge: | Recent studies have focused on whether large language models are capable of planning or executing plans. |
| Approach: | They propose an abductive reasoning task using wikiHow to test the effectiveness of small models over large models. |
| Outcome: | The proposed task demonstrates the effectiveness of small models over large models in most scenarios. |
Cetvel: A Unified Benchmark for Evaluating Language Understanding, Generation and Cultural Capacity of LLMs for Turkish (2026.eacl-long)
Copied to clipboard
| Challenge: | Existing Turkish benchmarks lack task diversity or culturally relevant content . Cetvel combines a broad range of discriminative and generative tasks . |
| Approach: | They propose a benchmark to evaluate large language models in Turkish . Cetvel combines a broad range of discriminative and generative tasks . they find that Turkish-centric instruction-tuned models generally underperform . |
| Outcome: | The proposed benchmark covers 23 tasks grouped into seven categories . it shows that Turkish-centric instruction-tuned models underperform relative to multilingual or general-purpose models despite being tailored for the language. |
A Zero-Shot Open-Vocabulary Pipeline for Dialogue Understanding (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing approaches to DST are limited by their computational resources or lack flexibility to adapt to new slots. |
| Approach: | They propose a system that integrates domain classification and DST in a single pipeline and uses self-refining prompts to adapt dynamically. |
| Outcome: | The proposed system improves on existing methods on multiWOZ datasets and provides 20% better Joint Goal Accuracy (JGA) over existing methods with 90% fewer requests to the LLM API. |
Character-Level Models versus Morphology in Semantic Role Labeling (P18-1)
Copied to clipboard
| Challenge: | Character-level models are used for high-level semantic analysis tasks such as semantic role labeling. |
| Approach: | They train character-level models that use word, character and morphology level information . they analyze how performance of characters compare to words and a variety of morphological typologies . |
| Outcome: | The results shed light on important characteristics of character-level models and their semantic capability. |
Text Processing Like Humans Do: Visually Attacking and Shielding NLP Systems (N19-1)
Copied to clipboard
Steffen Eger, Gözde Gül Şahin, Andreas Rücklé, Ji-Ung Lee, Claudia Schulz, Mohsen Mesgar, Krishnkant Swarnkar, Edwin Simpson, Iryna Gurevych
| Challenge: | Recent studies show that visual similarity can play a decisive role in assessing the meaning of characters. |
| Approach: | They investigate the impact of visual adversarial attacks on current NLP systems . they explore three shielding methods that significantly improve the robustness of the models . |
| Outcome: | The proposed methods improve performance but still fall behind non-attack scenarios. |
GECTurk WEB: An Explainable Online Platform for Turkish Grammatical Error Detection and Correction (2025.coling-demos)
Copied to clipboard
| Challenge: | GECTurk WEB is an open-source, web-based system that can detect and correct most common forms of Turkish writing errors. |
| Approach: | They propose a web-based system that detects and corrects most common errors in Turkish . it provides an easy-to-use tool for native speakers and second language learners . |
| Outcome: | The proposed system achieves 88,3 system usability score and is shown to help learn/remember a grammatical rule. |
PuzzLing Machines: A Challenge on Learning From Small Data (2020.acl-main)
Copied to clipboard
| Challenge: | a benchmark dataset of 81 languages is released to test deep neural models' human-like reasoning and generalization skills. |
| Approach: | They propose a challenge on learning from small data using Rosetta Stone puzzles from Linguistic Olympiads for high school students. |
| Outcome: | The proposed benchmark consists of Rosetta Stone puzzles from Linguistic Olympiads for high school students. |