SMARTER: A Data-efficient Framework to Improve Toxicity Detection with Explanation via Self-augmenting Large Language Models

arXiv:2509.15174v3 Announce Type: replace-cross Abstract: WARNING: This paper contains examples of offensive materials. To address the proliferation of toxic content on social media, we introduce SMARTER, we introduce SMARTER, a data-efficient two-stage framework for explainable content moderation using Large Language Models (LLMs). In Stage 1, we leverage LLMs' own outputs to generate synthetic explanations for both correct and incorrect labels, enabling alignment via preference optimization with minimal human supervision. In Stage 2, we refine explanation quality through cross-model training, allowing weaker models to align stylistically and semantically with stronger ones. Experiments on three benchmark tasks -- HateXplain, Latent Hate, and Implicit Hate -- demonstrate that SMARTER enables LLMs to achieve up to a 13% macro-F1 improvement over standard few-shot baselines while using only a fraction of the full training data. Our framework offers a scalable strategy for low-resource settings by harnessing LLMs' self-improving capabilities for both classification and explanation.

Related

• MERMAID: Memory-Enhanced Retrieval and Reasoning with Multi-Agent Iterative Knowledge Grounding for Veracity Assessment
• Why AI-Generated Text Detection Fails: Evidence from Explainable AI Beyond Benchmark Accuracy
• Safe-SAIL: Towards a Fine-grained Safety Landscape of Large Language Models via Sparse Autoencoder Interpretation Framework
• SelfGrader: Stable Jailbreak Detection for Large Language Models using Token-Level Logits

Source: arXiv cs.AI | 2026-04-23