Papers by Huy Nghiem
SMARTER: A Data-efficient Framework to Improve Toxicity Detection with Explanation via Self-augmenting Large Language Models (2026.acl-long)
Copied to clipboard
| Challenge: | Toxic content encompasses a wide spectrum of terminologies whose definitions vary by platform. |
| Approach: | They propose a 2-stage framework for explainable content moderation using Large Language Models (LLMs) they leverage LLMs’ own outputs to generate synthetic explanations for correct and incorrect labels . they refine explanation quality through cross-model training, allowing weaker models to align with stronger ones. |
| Outcome: | Experiments on 3 benchmarks show that the proposed framework achieves 13% macro-F1 improvement over few-shot baselines using only 6-57% of training data. |
PEDANTS: Cheap but Effective and Interpretable Answer Equivalence (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Current short-form QA evaluations lack diverse styles of evaluation data and rely on expensive and slow LLMs. |
| Approach: | They propose a rubric for machine QA that is more stable than an exact match and neural methods. |
| Outcome: | The proposed evaluations improve on the existing short-form QA evaluations using the Trivia community. |
Can You Make It Sound Like You? Post-Editing LLM-Generated Text for Personal Style (2026.acl-long)
Copied to clipboard
| Challenge: | Despite the growing use of large language models for writing tasks, it remains unclear whether users can effectively reshape LLM-generated text to reflect their personal style. |
| Approach: | They conduct an online study in which participants post-edit LLM-generated drafts for writing tasks where personal style matters to them. |
| Outcome: | The results show that post-editing increases stylistic similarity to unassisted writing and reduces similarity with fully LLM-generated output. |
HateCOT: An Explanation-Enhanced Dataset for Generalizable Offensive Speech Detection via Large Language Models (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Social media has amplified the propagation of hateful sentiments, highlighting the contested nature of "offensive content" research shows that "of offensive content" is still a contested construct due to varying definitions and labeling. |
| Approach: | They propose a dataset that features human-curated explanations for offensive content in English . they show that HateCOT pretraining improves performance of open-source LLMs . |
| Outcome: | The proposed model improves on three benchmark datasets for offensive content detection . the model improve the quality of its explanations, as confirmed by the human evaluation . |
VietMix: A Naturally-Occurring Parallel Corpus and Augmentation Framework for Vietnamese-English Code-Mixed Machine Translation (2026.eacl-long)
Copied to clipboard
| Challenge: | Existing approaches to machine translation (MT) systems degrade when faced with code-mixed text. |
| Approach: | They propose a system that can augment Vietnamese-English code-mixed text with iterative fine-tuning and targeted filtering. |
| Outcome: | The proposed framework outperforms strong back-translation baselines and improves zero-shot models by up to +11.9 points. |
“You Gotta be a Doctor, Lin” : An Investigation of Name-Based Bias of Large Language Models in Employment Recommendations (2024.emnlp-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have demonstrated racial and gender biases in various applications. |
| Approach: | They use Large Language Models to simulate hiring decisions and salary recommendations for candidates with 320 first names that strongly signal their race and gender, across over 750,000 prompts. |
| Outcome: | The proposed models favor candidates with White female-sounding names over other demographic groups across 40 occupations. |
“Define Your Terms” : Enhancing Efficient Offensive Speech Classification with Definition (2024.eacl-long)
Copied to clipboard
| Challenge: | Multiple studies have proposed various semantically related yet subtle distinct categories of offensive speech. |
| Approach: | They propose a meta-learning architecture that incorporates the input’s label and definition for classification via Prototypical Network. |
| Outcome: | The proposed model achieves 75% of the maximal F1-score while using less than 10% of the available training data across 4 datasets. |
‘Rich Dad, Poor Lad’: How do Large Language Models Contextualize Socioeconomic Factors in College Admission ? (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large Language Models are increasingly involved in high-stakes domains, yet how they reason about socially sensitive decisions remains underexplored. |
| Approach: | They propose a dual-process audit framework to probe LLMs’ reasoning behaviors in sensitive applications using a synthetic dataset of 30,000 applicant profiles grounded in real-world correlations. |
| Outcome: | The proposed framework exploits a synthetic dataset of 30,000 applicant profiles grounded in real-world correlations to probe LLMs' reasoning behaviors in sensitive applications. |