arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2601.21722cs.CLcs.AI
收藏

增强语言模型以实现稳健的绿色洗脑检测

Enhancing Language Models for Robust Greenwashing Detection

National University of Singapore(新加坡国立大学);Massachusetts Institute of Technology(麻省理工学院);Nanyang Technological University(南洋理工大学)

Neil Heinrich Braun, Keane Ong, Rui Mao, Erik Cambria, Gianmarco Mengaldo

AI总结:

本文提出了一种参数高效框架,通过结合对比学习和顺序排名目标来增强语言模型在绿色洗脑检测中的鲁棒性。

摘要:

可持续性报告对于ESG评估至关重要,但绿色洗脑和模糊声明常常削弱其可靠性。现有NLP模型缺乏对这些做法的鲁棒性,通常依赖于表面模式,泛化能力差。我们提出了一种参数高效框架,通过结合对比学习和顺序排名目标来构建LLM潜在空间,以捕捉具体行动与模糊声明之间的等级差异。我们的方法结合了门控特征调节以过滤披露噪声,并利用MetaGradNorm来稳定多目标优化。在跨类别设置中的实验表明,我们的方法在标准基线之上表现出更强的鲁棒性,同时揭示了表示刚性和泛化之间的权衡。

英文摘要:

Sustainability reports are critical for ESG assessment, yet greenwashing and vague claims often undermine their reliability. Existing NLP models lack robustness to these practices, typically relying on surface-level patterns that generalize poorly. We propose a parameter-efficient framework that structures LLM latent spaces by combining contrastive learning with an ordinal ranking objective to capture graded distinctions between concrete actions and ambiguous claims. Our approach incorporates gated feature modulation to filter disclosure noise and utilizes MetaGradNorm to stabilize multi-objective optimization. Experiments in cross-category settings demonstrate superior robustness over standard baselines while revealing a trade-off between representational rigidity and generalization.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑