Evaluating Proactive Risk Awareness of Large Language Models
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866912923305115648 |
|---|---|
| author | Luo, Xuan Chen, Yubin Hou, Zhiyu Yu, Linpu Tu, Geng Li, Jing Xu, Ruifeng |
| author_facet | Luo, Xuan Chen, Yubin Hou, Zhiyu Yu, Linpu Tu, Geng Li, Jing Xu, Ruifeng |
| contents | As large language models (LLMs) are increasingly embedded in everyday decision-making, their safety responsibilities extend beyond reacting to explicit harmful intent toward anticipating unintended but consequential risks. In this work, we introduce a proactive risk awareness evaluation framework that measures whether LLMs can anticipate potential harms and provide warnings before damage occurs. We construct the Butterfly dataset to instantiate this framework in the environmental and ecological domain. It contains 1,094 queries that simulate ordinary solution-seeking activities whose responses may induce latent ecological impact. Through experiments across five widely used LLMs, we analyze the effects of response length, languages, and modality. Experimental results reveal consistent, significant declines in proactive awareness under length-restricted responses, cross-lingual similarities, and persistent blind spots in (multimodal) species protection. These findings highlight a critical gap between current safety alignment and the requirements of real-world ecological responsibility, underscoring the need for proactive safeguards in LLM deployment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_20976 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Evaluating Proactive Risk Awareness of Large Language Models Luo, Xuan Chen, Yubin Hou, Zhiyu Yu, Linpu Tu, Geng Li, Jing Xu, Ruifeng Computation and Language Computers and Society As large language models (LLMs) are increasingly embedded in everyday decision-making, their safety responsibilities extend beyond reacting to explicit harmful intent toward anticipating unintended but consequential risks. In this work, we introduce a proactive risk awareness evaluation framework that measures whether LLMs can anticipate potential harms and provide warnings before damage occurs. We construct the Butterfly dataset to instantiate this framework in the environmental and ecological domain. It contains 1,094 queries that simulate ordinary solution-seeking activities whose responses may induce latent ecological impact. Through experiments across five widely used LLMs, we analyze the effects of response length, languages, and modality. Experimental results reveal consistent, significant declines in proactive awareness under length-restricted responses, cross-lingual similarities, and persistent blind spots in (multimodal) species protection. These findings highlight a critical gap between current safety alignment and the requirements of real-world ecological responsibility, underscoring the need for proactive safeguards in LLM deployment. |
| title | Evaluating Proactive Risk Awareness of Large Language Models |
| topic | Computation and Language Computers and Society |
| url | https://arxiv.org/abs/2602.20976 |