Evaluating LLM-Generated Versus Human-Authored Responses in Role-Play Dialogues
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909832083144704 |
|---|---|
| author | Lu, Dongxu Jeuring, Johan Gatt, Albert |
| author_facet | Lu, Dongxu Jeuring, Johan Gatt, Albert |
| contents | Evaluating large language models (LLMs) in long-form, knowledge-grounded role-play dialogues remains challenging. This study compares LLM-generated and human-authored responses in multi-turn professional training simulations through human evaluation ($N=38$) and automated LLM-as-a-judge assessment. Human evaluation revealed significant degradation in LLM-generated response quality across turns, particularly in naturalness, context maintenance and overall quality, while human-authored responses progressively improved. In line with this finding, participants also indicated a consistent preference for human-authored dialogue. These human judgements were validated by our automated LLM-as-a-judge evaluation, where Gemini 2.0 Flash achieved strong alignment with human evaluators on both zero-shot pairwise preference and stochastic 6-shot construct ratings, confirming the widening quality gap between LLM and human responses over time. Our work contributes a multi-turn benchmark exposing LLM degradation in knowledge-grounded role-play dialogues and provides a validated hybrid evaluation framework to guide the reliable integration of LLMs in training simulations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_17694 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Evaluating LLM-Generated Versus Human-Authored Responses in Role-Play Dialogues Lu, Dongxu Jeuring, Johan Gatt, Albert Computation and Language Artificial Intelligence Evaluating large language models (LLMs) in long-form, knowledge-grounded role-play dialogues remains challenging. This study compares LLM-generated and human-authored responses in multi-turn professional training simulations through human evaluation ($N=38$) and automated LLM-as-a-judge assessment. Human evaluation revealed significant degradation in LLM-generated response quality across turns, particularly in naturalness, context maintenance and overall quality, while human-authored responses progressively improved. In line with this finding, participants also indicated a consistent preference for human-authored dialogue. These human judgements were validated by our automated LLM-as-a-judge evaluation, where Gemini 2.0 Flash achieved strong alignment with human evaluators on both zero-shot pairwise preference and stochastic 6-shot construct ratings, confirming the widening quality gap between LLM and human responses over time. Our work contributes a multi-turn benchmark exposing LLM degradation in knowledge-grounded role-play dialogues and provides a validated hybrid evaluation framework to guide the reliable integration of LLMs in training simulations. |
| title | Evaluating LLM-Generated Versus Human-Authored Responses in Role-Play Dialogues |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2509.17694 |