Evaluating LLM Safety Across Child Development Stages: A Simulated Agent Approach
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918293246312448 |
|---|---|
| author | Murali, Abhejay Afroogh, Saleh Chen, Kevin Atkinson, David Dhurandhar, Amit Jiao, Junfeng |
| author_facet | Murali, Abhejay Afroogh, Saleh Chen, Kevin Atkinson, David Dhurandhar, Amit Jiao, Junfeng |
| contents | Current safety alignment for Large Language Models (LLMs) implicitly optimizes for a "modal adult user," leaving models vulnerable to distributional shifts in user cognition. We present ChildSafe, a benchmark that quantifies alignment robustness under cognitive shifts corresponding to four developmental stages. Unlike static persona-based evaluations, we introduce a parametric cognitive simulation approach, formalizing developmental stages as hyperparameter constraints (e.g., volatility, context horizon) to generate out-of-distribution interaction traces. We validate these agents against ground-truth human linguistic data (CHILDES) and deploy them across 1,200 multi-turn interactions. Our results reveal a systematic alignment generalization gap: state-of-the-art models exhibit up to 11.5% performance degradation when interacting with early-childhood agents compared to standard baselines. We provide the research community with the validated agent artifacts and evaluation protocols to facilitate robust alignment testing against non-adversarial, cognitively diverse populations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_05484 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Evaluating LLM Safety Across Child Development Stages: A Simulated Agent Approach Murali, Abhejay Afroogh, Saleh Chen, Kevin Atkinson, David Dhurandhar, Amit Jiao, Junfeng Computers and Society Current safety alignment for Large Language Models (LLMs) implicitly optimizes for a "modal adult user," leaving models vulnerable to distributional shifts in user cognition. We present ChildSafe, a benchmark that quantifies alignment robustness under cognitive shifts corresponding to four developmental stages. Unlike static persona-based evaluations, we introduce a parametric cognitive simulation approach, formalizing developmental stages as hyperparameter constraints (e.g., volatility, context horizon) to generate out-of-distribution interaction traces. We validate these agents against ground-truth human linguistic data (CHILDES) and deploy them across 1,200 multi-turn interactions. Our results reveal a systematic alignment generalization gap: state-of-the-art models exhibit up to 11.5% performance degradation when interacting with early-childhood agents compared to standard baselines. We provide the research community with the validated agent artifacts and evaluation protocols to facilitate robust alignment testing against non-adversarial, cognitively diverse populations. |
| title | Evaluating LLM Safety Across Child Development Stages: A Simulated Agent Approach |
| topic | Computers and Society |
| url | https://arxiv.org/abs/2510.05484 |