Probing Social Identity Bias in Chinese LLMs with Gendered Pronouns and Social Groups

• URL: http://arxiv.org/abs/2510.06974v1
• Date: Wed, 08 Oct 2025 13:00:12 GMT
• Title: Probing Social Identity Bias in Chinese LLMs with Gendered Pronouns and Social Groups
• Abstract summary: Large language models (LLMs) are increasingly deployed in user-facing applications, raising concerns about their potential to reflect and amplify social biases.<n>We investigate social identity framing in Chinese LLMs using Mandarin-specific prompts, evaluating responses to ingroup ("We") and outgroup ("They") framings, and extending the setting to 240 social groups salient in the Chinese context.<n>Across models, we observe systematic ingroup-positive and outgroup-negative tendencies, which are not confined to synthetic prompts but also appear in naturalistic dialogue.
• Score: 6.097454027594837
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models (LLMs) are increasingly deployed in user-facing applications, raising concerns about their potential to reflect and amplify social biases. We investigate social identity framing in Chinese LLMs using Mandarin-specific prompts across ten representative Chinese LLMs, evaluating responses to ingroup ("We") and outgroup ("They") framings, and extending the setting to 240 social groups salient in the Chinese context. To complement controlled experiments, we further analyze Chinese-language conversations from a corpus of real interactions between users and chatbots. Across models, we observe systematic ingroup-positive and outgroup-negative tendencies, which are not confined to synthetic prompts but also appear in naturalistic dialogue, indicating that bias dynamics might strengthen in real interactions. Our study provides a language-aware evaluation framework for Chinese LLMs, demonstrating that social identity biases documented in English generalize cross-linguistically and intensify in user-facing contexts.

Related papers

• Persuasive and Compliant Tendencies Predict Group Decision-Making in Humans and Language Models [65.21691850622413]
  Large language models (LLMs) are increasingly involved in group decision-making with humans.<n>We introduce DecisionQE, a questionnaire-based framework for measuring each model's persuasive and compliant tendencies.<n>We show that stronger persuasive tendency does not significantly improve group outcomes, whereas compliant-oriented models show more stable advantages in cooperation.
  arXiv  Detail & Related papers  (2026-08-08T15:50:56Z)
• The Language You Ask In: Language-Conditioned Ideological Divergence in LLM Analysis of Contested Political Documents [0.8528384027684192]
  This study presents an experimental comparison of large language models (LLMs) with prompts in Russian and Ukrainian.<n>Despite identical source material and parallel query structures, the resulting analyses varied substantially in rhetorical positioning, ideological orientation, and interpretive conclusions.<n>The Russian-language output echoed narratives common in Russian state discourse, characterizing civil society actors as illegitimate elites undermining democratic mandates.<n>The Ukrainian-language output adopted vocabulary characteristic of Western liberal-democratic political science, treating the same actors as legitimate stakeholders within democratic contestation.
  arXiv  Detail & Related papers  (2026-01-17T21:00:36Z)
• Social Simulations with Large Language Model Risk Utopian Illusion [61.358959720048354]
  We introduce a systematic framework for analyzing large language models' behavior in social simulation.<n>Our approach simulates multi-agent interactions through chatroom-style conversations and analyzes them across five linguistic dimensions.<n>Our findings reveal that LLMs do not faithfully reproduce genuine human behavior but instead reflect overly idealized versions of it.
  arXiv  Detail & Related papers  (2025-10-24T06:08:41Z)
• Are they lovers or friends? Evaluating LLMs' Social Reasoning in English and Korean Dialogues [32.86600072956612]
  We introduce SCRIPTS, a 1k-dialogue dataset in English and Korean, sourced from movie scripts.<n>The task involves evaluating models' social reasoning capability to infer the interpersonal relationships between speakers in each dialogue.<n>We find that thinking models and chain-of-thought prompting, effective for general reasoning, provide minimal benefits for social reasoning and occasionally amplify social biases.
  arXiv  Detail & Related papers  (2025-10-21T19:12:47Z)
• Toxicity Red-Teaming: Benchmarking LLM Safety in Singapore's Low-Resource Languages [57.059267233093465]
  Large Language Models (LLMs) have transformed natural language processing, but their safety mechanisms remain under-explored in low-resource, multilingual settings.<n>We introduce textsfSGToxicGuard, a novel dataset and evaluation framework for benchmarking LLM safety in Singapore's diverse linguistic context.<n>We conduct extensive experiments with state-of-the-art multilingual LLMs, and the results uncover critical gaps in their safety guardrails.
  arXiv  Detail & Related papers  (2025-09-18T08:14:34Z)
• Simulating Identity, Propagating Bias: Abstraction and Stereotypes in LLM-Generated Text [8.988787218035414]
  We investigate whether persona-prompting leads to different levels of linguistic abstraction when generating short texts linking socio-demographic categories with stereotypical or non-stereotypical attributes.<n>Our results confirm criticisms about the ecology of personas as representative of socio-demographic groups and raise concerns about the risk of propagating stereotypes.
  arXiv  Detail & Related papers  (2025-09-10T10:49:21Z)
• Evaluating Prompt-Driven Chinese Large Language Models: The Influence of Persona Assignment on Stereotypes and Safeguards [3.1308581258317485]
  We analyze how persona assignment influences refusal behavior and response toxicity in Qwen, a widely-used Chinese language model.<n>Our study reveals significant gender biases in refusal rates and demonstrates that certain negative personas can amplify toxicity toward Chinese social groups by up to 60-fold.<n>To mitigate this toxicity, we propose an innovative multi-model feedback strategy, employing iterative interactions between Qwen and an external evaluator.
  arXiv  Detail & Related papers  (2025-06-05T12:47:21Z)
• Characterizing Bias: Benchmarking Large Language Models in Simplified versus Traditional Chinese [52.98034458924209]
  This study investigates whether Large Language Models exhibit differential performance when prompted in two variants of written Chinese.<n>We design two benchmark tasks that reflect real-world scenarios: regional term choice and regional name choice.<n>Our analyses indicate that biases in LLM responses are dependent on both the task and prompting language.
  arXiv  Detail & Related papers  (2025-05-28T17:56:49Z)
• LLMs syntactically adapt their language use to their conversational partner [58.92470092706263]
  It has been frequently observed that human speakers align their language use with each other during conversations.<n>We construct a corpus of conversations between large language models (LLMs) and find that two LLM agents end up making more similar syntactic choices as conversations go on.
  arXiv  Detail & Related papers  (2025-03-10T15:37:07Z)
• Large Language Models Reflect the Ideology of their Creators [71.65505524599888]
  Large language models (LLMs) are trained on vast amounts of data to generate natural language.<n>This paper shows that the ideological stance of an LLM appears to reflect the worldview of its creators.
  arXiv  Detail & Related papers  (2024-10-24T04:02:30Z)
• Are Social Sentiments Inherent in LLMs? An Empirical Study on Extraction of Inter-demographic Sentiments [14.143299702954023]
  This study focuses on social groups defined in terms of nationality, religion, and race/ethnicity. We input questions regarding sentiments from one group to another into LLMs, apply sentiment analysis to the responses, and compare the results with social surveys.
  arXiv  Detail & Related papers  (2024-08-08T08:13:25Z)
• Generative Language Models Exhibit Social Identity Biases [17.307292780517653]
  We investigate whether ingroup solidarity and outgroup hostility, fundamental social identity biases, are present in 56 large language models. We find that almost all foundational language models and some instruction fine-tuned models exhibit clear ingroup-positive and outgroup-negative associations when prompted to complete sentences. Our findings suggest that modern language models exhibit fundamental social identity biases to a similar degree as humans.
  arXiv  Detail & Related papers  (2023-10-24T13:17:40Z)
• CMMLU: Measuring massive multitask language understanding in Chinese [133.70911295934746]
  This paper introduces a comprehensive Chinese benchmark that covers various subjects, including natural science, social sciences, engineering, and humanities. CMMLU fills the gap in evaluating the knowledge and reasoning capabilities of large language models within the Chinese context.
  arXiv  Detail & Related papers  (2023-06-15T15:49:51Z)
• Cue-CoT: Chain-of-thought Prompting for Responding to In-depth Dialogue Questions with LLMs [59.74002011562726]
  We propose a novel linguistic cue-based chain-of-thoughts (textitCue-CoT) to provide a more personalized and engaging response. We build a benchmark with in-depth dialogue questions, consisting of 6 datasets in both Chinese and English. Empirical results demonstrate our proposed textitCue-CoT method outperforms standard prompting methods in terms of both textithelpfulness and textitacceptability on all datasets.
  arXiv  Detail & Related papers  (2023-05-19T16:27:43Z)

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.