Benchmarking Educational LLMs with Analytics: A Case Study on Gender Bias in Feedback

• URL: http://arxiv.org/abs/2511.08225v1
• Date: Wed, 12 Nov 2025 01:47:34 GMT
• Title: Benchmarking Educational LLMs with Analytics: A Case Study on Gender Bias in Feedback
• Authors: Yishan Du, Conrad Borchers, Mutlu Cukurova,
• Abstract summary: This article presents an embedding-based benchmarking framework to detect bias in large language models (LLMs)<n>Using 600 authentic student essays from the AES 2.0 corpus, we constructed controlled counterfactuals along two dimensions.<n>We investigated six representative LLMs (i.e. GPT-5 mini, GPT-4o mini, DeepSeek-R1, DeepSeek-R1-Qwen, Gemini 2.5 Pro, Llama-3-8B)<n>In all models, implicit manipulations reliably induced larger semantic shifts for male-female counterfactuals than for female-male.
• Score: 0.06951558137930917
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: As teachers increasingly turn to GenAI in their educational practice, we need robust methods to benchmark large language models (LLMs) for pedagogical purposes. This article presents an embedding-based benchmarking framework to detect bias in LLMs in the context of formative feedback. Using 600 authentic student essays from the AES 2.0 corpus, we constructed controlled counterfactuals along two dimensions: (i) implicit cues via lexicon-based swaps of gendered terms within essays, and (ii) explicit cues via gendered author background in the prompt. We investigated six representative LLMs (i.e. GPT-5 mini, GPT-4o mini, DeepSeek-R1, DeepSeek-R1-Qwen, Gemini 2.5 Pro, Llama-3-8B). We first quantified the response divergence with cosine and Euclidean distances over sentence embeddings, then assessed significance via permutation tests, and finally, visualised structure using dimensionality reduction. In all models, implicit manipulations reliably induced larger semantic shifts for male-female counterfactuals than for female-male. Only the GPT and Llama models showed sensitivity to explicit gender cues. These findings show that even state-of-the-art LLMs exhibit asymmetric semantic responses to gender substitutions, suggesting persistent gender biases in feedback they provide learners. Qualitative analyses further revealed consistent linguistic differences (e.g., more autonomy-supportive feedback under male cues vs. more controlling feedback under female cues). We discuss implications for fairness auditing of pedagogical GenAI, propose reporting standards for counterfactual evaluation in learning analytics, and outline practical guidance for prompt design and deployment to safeguard equitable feedback.

Related papers

• Exploring Gender Bias in Large Language Models: An In-depth Dive into the German Language [21.87606488958834]
  We present five German datasets for gender bias evaluation in large language models (LLMs)<n>The datasets are grounded in well-established concepts of gender bias and are accessible through multiple methodologies.<n>Our findings, reported for eight multilingual LLM models, reveal unique challenges associated with gender bias in German.
  arXiv  Detail & Related papers  (2025-07-22T13:09:41Z)
• Detection, Classification, and Mitigation of Gender Bias in Large Language Models [6.762310697831219]
  We investigate how to enhance the capabilities of large language models (LLMs) in gender bias detection, classification, and mitigation.<n>We adopt reinforcement learning, chain-of-thoughts reasoning, and supervised fine-tuning to handle different Subtasks.<n>Our approach ranked first across all three subtasks of the NLPCC 2025 Shared Task 7.
  arXiv  Detail & Related papers  (2025-06-14T14:53:25Z)
• The Root Shapes the Fruit: On the Persistence of Gender-Exclusive Harms in Aligned Language Models [91.86718720024825]
  We center transgender, nonbinary, and other gender-diverse identities to investigate how alignment procedures interact with pre-existing gender-diverse bias.<n>Our findings reveal that DPO-aligned models are particularly sensitive to supervised finetuning.<n>We conclude with recommendations tailored to DPO and broader alignment practices.
  arXiv  Detail & Related papers  (2024-11-06T06:50:50Z)
• GenderCARE: A Comprehensive Framework for Assessing and Reducing Gender Bias in Large Language Models [73.23743278545321]
  Large language models (LLMs) have exhibited remarkable capabilities in natural language generation, but have also been observed to magnify societal biases.<n>GenderCARE is a comprehensive framework that encompasses innovative Criteria, bias Assessment, Reduction techniques, and Evaluation metrics.
  arXiv  Detail & Related papers  (2024-08-22T15:35:46Z)
• Beyond Binary Gender: Evaluating Gender-Inclusive Machine Translation with Ambiguous Attitude Words [85.48043537327258]
  Existing machine translation gender bias evaluations are primarily focused on male and female genders. This study presents a benchmark AmbGIMT (Gender-Inclusive Machine Translation with Ambiguous attitude words) We propose a novel process to evaluate gender bias based on the Emotional Attitude Score (EAS), which is used to quantify ambiguous attitude words.
  arXiv  Detail & Related papers  (2024-07-23T08:13:51Z)
• GenderBias-\emph{VL}: Benchmarking Gender Bias in Vision Language Models via Counterfactual Probing [72.0343083866144]
  This paper introduces the GenderBias-emphVL benchmark to evaluate occupation-related gender bias in Large Vision-Language Models. Using our benchmark, we extensively evaluate 15 commonly used open-source LVLMs and state-of-the-art commercial APIs. Our findings reveal widespread gender biases in existing LVLMs.
  arXiv  Detail & Related papers  (2024-06-30T05:55:15Z)
• Unraveling Downstream Gender Bias from Large Language Models: A Study on AI Educational Writing Assistance [13.276943737418417]
  Large Language Models (LLMs) are increasingly utilized in educational tasks such as providing writing suggestions to students. Previous studies have investigated bias in models and data representations separately. We investigate how bias transfers through an AI writing support pipeline.
  arXiv  Detail & Related papers  (2023-11-06T18:01:34Z)
• Probing Explicit and Implicit Gender Bias through LLM Conditional Text Generation [64.79319733514266]
  Large Language Models (LLMs) can generate biased and toxic responses. We propose a conditional text generation mechanism without the need for predefined gender phrases and stereotypes.
  arXiv  Detail & Related papers  (2023-11-01T05:31:46Z)
• Towards Understanding Gender-Seniority Compound Bias in Natural Language Generation [64.65911758042914]
  We investigate how seniority impacts the degree of gender bias exhibited in pretrained neural generation models. Our results show that GPT-2 amplifies bias by considering women as junior and men as senior more often than the ground truth in both domains. These results suggest that NLP applications built using GPT-2 may harm women in professional capacities.
  arXiv  Detail & Related papers  (2022-05-19T20:05:02Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.