Assessing Dialect Fairness and Robustness of Large Language Models in Reasoning Tasks (2025.acl-long)
Copied to clipboard
Fangru Lin, Shaoguang Mao, Emanuele La Malfa, Valentin Hofmann, Adrian de Wynter, Xun Wang, Si-Qing Chen, Michael J. Wooldridge, Janet B. Pierrehumbert, Furu Wei
| Challenge: | a study aims to assess the fairness and robustness of Large Language Models in dialectal queries . speakers of "non-standard" dialects are known to experience implicit and explicit discrimination . |
| Approach: | They propose to use a benchmark to assess the fairness of large language models in dialects . they hire speakers with computer science backgrounds to rewrite seven popular benchmarks based on AAVE . |
| Outcome: | The proposed benchmarks show that most models show significant brittleness and unfairness to queries in AAVE. |
Similar Papers
Analyzing Dialectical Biases in LLMs for Knowledge and Reasoning Benchmarks (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Previous work has shown degraded performance of large language models for under-represented English dialects. |
| Approach: | They analyze the effects of typifying “standard” American English language questions as non-”standard” dialectal variants on multiple choice questions. |
| Outcome: | The results show that typifying “standard” American English language questions as non-”standard” dialectal variants can reduce performance 20% . |
EnDive: A Cross-Dialect Benchmark for Fairness and Performance in Large Language Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing benchmarks often overlook intra-language variations, leaving speakers of non-standard dialects underserved. |
| Approach: | EnDive evaluates seven state-of-the-art large language models across tasks . human evaluations confirm high translation quality, with average scores of at least 6.02/7 . |
| Outcome: | EnDive evaluates state-of-the-art large language models across language understanding, reasoning, mathematics, logic tasks. |
Rejected Dialects: Biases Against African American Language in Reward Models (2025.findings-naacl)
Copied to clipboard
| Challenge: | Preference alignment via reward models can introduce new biases, hindering reward models’ fairness and equity. |
| Approach: | They propose a framework for evaluating dialect biases in reward models and conduct a case study on biase . they compare reward models' preferences and behavior on paired White Mainstream English and machine-translated and human-written AAL corpora. |
| Outcome: | The proposed framework evaluates dialect biases in reward models and compares them with paired White Mainstream English (WME) and machine-translated and human-written AAL corpora. |
VALUE: Understanding Dialect Disparity in NLU (2022.acl-long)
Copied to clipboard
| Challenge: | English Natural Language Understanding systems outperform humans on benchmarks like GLUE and SuperGLUE, but they only use textbook Standard American English (SAE) . fewer studies have considered the effects of dialectal differences on performance . |
| Approach: | They propose a benchmark to evaluate the performance of English natural language understanding systems using a set of lexical and morphosyntactic transformation rules. |
| Outcome: | The proposed model outperforms humans on GLUE and SuperGLUE, but only on standard American English . the proposed model recruits fluent speakers of African American vernacular english to validate each feature transformation . |
FLEX: A Benchmark for Evaluating Robustness of Fairness in Large Language Models (2025.findings-naacl)
Copied to clipboard
| Challenge: | Existing safety evaluations may overlook the inherent weaknesses of Large Language Models, despite their benefits. |
| Approach: | They propose a benchmark to evaluate the robustness of Large Language Models under extreme conditions. |
| Outcome: | The proposed approach evaluates the fairness of large language models under extreme conditions. |
Exploring the Reliability of Large Language Models as Customized Evaluators for Diverse NLP Tasks (2025.coling-main)
Copied to clipboard
| Challenge: | Existing work uses large language models (LLMs) to evaluate natural language process tasks, but there are shortcomings in current LLMs. |
| Approach: | They examine the alignment between LLM evaluators and human annotators by comparing conventional and alignment tasks with different evaluation criteria. |
| Outcome: | The proposed models excel in general criteria, such as fluency, but face challenges with complex criteria, including numerical reasoning. |
Evaluation of African American Language Bias in Natural Language Generation (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies have shown that large language generation models disadvantaging African American Language (AAL) can be biased for certain language varieties, but there is little research on the impact of these biases on other languages. |
| Approach: | They evaluate how well LLMs understand African American Language (AAL) in comparison to white Mainstream English (WME) using a dataset of AAL texts from a variety of regions and contexts, they find dialectal bias in six pre-trained LLM. |
| Outcome: | The proposed models understand African American language in comparison to white mainstream English (WME) the proposed models have performance gaps on two tasks that are not matched by the model. |
Toward Robust Evaluation for Multilingual Grammatical Error Correction: Can Large Language Models Replace Human References? (2026.acl-long)
Copied to clipboard
| Challenge: | Prior work has shown that using aclosest-gold reference yields more accurate performance estimates, but producing such references for each system individually is costly. |
| Approach: | They propose a method for generating closest-gold references by prompting a large language model with system outputs and a standard reference-based evaluations show weak or no correlation. |
| Outcome: | The proposed method outperforms state-of-the-art models on 14 languages across 14 benchmarks. |
Large Language Models are not Fair Evaluators (2024.acl-long)
Copied to clipboard
Peiyi Wang, Lei Li, Liang Chen, Zefan Cai, Dawei Zhu, Binghuai Lin, Yunbo Cao, Lingpeng Kong, Qi Liu, Tianyu Liu, Zhifang Sui
| Challenge: | Existing evaluation frameworks that use large language models as referees are insufficient for accurately assessing their alignment with human intent. |
| Approach: | They propose a calibration framework to address positional bias in large language models as evaluators by manually annotating the “win/tie/lose” outcomes of responses from ChatGPT and Vicuna-13B in the Vicun A Benchmark’s question prompt. |
| Outcome: | The proposed framework alleviates evaluation bias, resulting in closer alignment with human judgments. |
Do Large Language Models Rank Fairly? An Empirical Study on the Fairness of LLMs as Rankers (2024.naacl-long)
Copied to clipboard
| Challenge: | Recent studies have shown that Large Language Models (LLMs) are more efficient in natural language understanding tasks. |
| Approach: | They evaluate large language models (LLMs) using a TREC Fair Ranking dataset . they assess fairness from both user and content perspectives . |
| Outcome: | The proposed model outperforms the existing models in the fair ranking task. |