JurEE not Judges: safeguarding llm interactions with small, specialised Encoder Ensembles
This work addresses content moderation for customer-facing chatbots, offering an incremental improvement over existing LLM-as-Judge methods by enhancing generalization and interpretability.
The paper tackles the problem of safeguarding AI-user interactions in LLM-based systems by introducing JurEE, an ensemble of encoder-only transformer models that provide probabilistic risk estimates, significantly outperforming baseline models in accuracy, speed, and cost-efficiency on benchmarks like the OpenAI Moderation Dataset and ToxicChat.
We introduce JurEE, an ensemble of efficient, encoder-only transformer models designed to strengthen safeguards in AI-User interactions within LLM-based systems. Unlike existing LLM-as-Judge methods, which often struggle with generalization across risk taxonomies and only provide textual outputs, JurEE offers probabilistic risk estimates across a wide range of prevalent risks. Our approach leverages diverse data sources and employs progressive synthetic data generation techniques, including LLM-assisted augmentation, to enhance model robustness and performance. We create an in-house benchmark comprising of other reputable benchmarks such as the OpenAI Moderation Dataset and ToxicChat, where we find JurEE significantly outperforms baseline models, demonstrating superior accuracy, speed, and cost-efficiency. This makes it particularly suitable for applications requiring stringent content moderation, such as customer-facing chatbots. The encoder-ensemble's modular design allows users to set tailored risk thresholds, enhancing its versatility across various safety-related applications. JurEE's collective decision-making process, where each specialized encoder model contributes to the final output, not only improves predictive accuracy but also enhances interpretability. This approach provides a more efficient, performant, and economical alternative to traditional LLMs for large-scale implementations requiring robust content moderation.