Enhancing Language Models for Robust Greenwashing Detection
This work addresses greenwashing detection for ESG assessment, representing an incremental improvement with a novel method for a known bottleneck.
The paper tackles the problem of detecting greenwashing in sustainability reports by addressing the poor generalization of existing NLP models, achieving superior robustness in cross-category experiments.
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.