PulseAugur
LIVE 12:28:05
research · [2 sources] ·
1
research

LLMs generate realistic social networks, but prompt choices encode biases

A new study investigates how Large Language Models (LLMs) generate social networks, finding that factors like cultural framing, prompt language, and model scale significantly influence the outcomes. Researchers developed four tie-formation mechanisms and tested them across various conditions, revealing that political affiliation often dominates network formation, while prompt architecture can act as a sociological variable. The study also noted that while LLM-generated networks exhibit good clustering, they can encode demographic biases. AI

Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →

IMPACT Reveals how LLM outputs are shaped by prompt design, offering insights for researchers using LLMs in behavioral simulations.

RANK_REASON Academic paper detailing a study on LLM capabilities.

Read on arXiv cs.CL →

COVERAGE [2]

  1. arXiv cs.CL TIER_1 · Dalal Alharthi ·

    When Do LLMs Generate Realistic Social Networks? A Multi-Dimensional Study of Culture, Language, Scale, and Method

    Large language models (LLMs) are increasingly used as substitutes for human subjects in behavioral simulations, including synthetic social network generation. Yet it remains unclear how their relational outputs depend on prompt design, cultural framing, prompt language, and model…

  2. Hugging Face Daily Papers TIER_1 ·

    When Do LLMs Generate Realistic Social Networks? A Multi-Dimensional Study of Culture, Language, Scale, and Method

    Large language models (LLMs) are increasingly used as substitutes for human subjects in behavioral simulations, including synthetic social network generation. Yet it remains unclear how their relational outputs depend on prompt design, cultural framing, prompt language, and model…