# Active Learning for Robust and Representative LLM Generation in Safety-Critical Scenarios

arXiv: https://arxiv.org/abs/2410.11114 · HTML: https://mlsanity.com/paper/2410.11114/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 42 references that are in the map.
- First posted on arXiv 14 Oct 2024.

## Builds on (most cited first, 21 of 21)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [Gender Bias in Neural Natural Language Processing](https://mlsanity.com/paper/1807.11714/) (2018)
- [Active Learning by Acquiring Contrastive Examples](https://mlsanity.com/paper/2109.03764/) (2021)
- [Cold-start Active Learning through Self-supervised Language Modeling](https://mlsanity.com/paper/2010.09535/) (2020)
- [Increasing Diversity While Maintaining Accuracy: Text Data Generation with Large Language Models and Human Interventions](https://mlsanity.com/paper/2306.04140/) (2023)
- [Mitigating Language-Dependent Ethnic Bias in BERT](https://mlsanity.com/paper/2109.05704/) (2021)
- [Anticipating Safety Issues in E2E Conversational AI: Framework and Tooling](https://mlsanity.com/paper/2107.03451/) (2021)
- [On the Safety of Conversational Models: Taxonomy, Dataset, and Benchmark](https://mlsanity.com/paper/2110.08466/) (2021)
- [Practical Obstacles to Deploying Active Learning](https://mlsanity.com/paper/1807.04801/) (2018)
- [AART: AI-Assisted Red-Teaming with Diverse Data Generation for New LLM-powered Applications](https://mlsanity.com/paper/2311.08592/) (2023)
- [UmlsBERT: Clinical Domain Knowledge Augmentation of Contextual Embeddings Using the Unified Medical Language System Metathesaurus](https://mlsanity.com/paper/2010.10391/) (2020)
- [Balancing out Bias: Achieving Fairness Through Balanced Training](https://mlsanity.com/paper/2109.08253/) (2021)
- [On the Importance of Effectively Adapting Pretrained Language Models for Active Learning](https://mlsanity.com/paper/2104.08320/) (2021)
- [APPDIA: A Discourse-aware Transformer-based Style Transfer Model for Offensive Social Media Conversations](https://mlsanity.com/paper/2209.08207/) (2022)
- [Multi-task Active Learning for Pre-trained Transformer-based Models](https://mlsanity.com/paper/2208.05379/) (2022)
- [DisCGen: A Framework for Discourse-Informed Counterspeech Generation](https://mlsanity.com/paper/2311.18147/) (2023)
- [D-CALM: A Dynamic Clustering-based Active Learning Approach for Mitigating Bias](https://mlsanity.com/paper/2305.17013/) (2023)
- [Pre-trained Language Model Based Active Learning for Sentence Matching](https://mlsanity.com/paper/2010.05522/) (2020)

## Built on by (most cited first, 10 of 10)

- [The State of Multilingual LLM Safety Research: From Measuring the Language Gap to Mitigating It](https://mlsanity.com/paper/2505.24119/) (2025)
- [ProbeLLM: Automating Principled Diagnosis of LLM Failures](https://mlsanity.com/paper/2602.12966/) (2026)
- [Coherence-Driven Multimodal Safety Dialogue with Active Learning for Embodied Agents](https://mlsanity.com/paper/2410.14141/) (2024)
- [An Active Learning Framework for Inclusive Generation by Large Language Models](https://mlsanity.com/paper/2410.13641/) (2024)
- [Interpreting the Effects of Quantization on LLMs](https://mlsanity.com/paper/2508.16785/) (2025)
- [Automated Capability Evaluation of Foundation Models](https://mlsanity.com/paper/2505.17228/) (2025)
- [Reconstructing Biological Pathways by Applying Selective Incremental Learning to (Very) Small Language Models](https://mlsanity.com/paper/2507.04432/) (2025)
- [Moonworks Lunara Aesthetic Dataset](https://mlsanity.com/paper/2601.07941/) (2026)
- [Moonworks Lunara Aesthetic II: An Image Variation Dataset](https://mlsanity.com/paper/2602.01666/) (2026)
- [Moonworks Lunara: Modeling Artistic Intelligence](https://mlsanity.com/paper/2609.22272/) (2026)

## Abstract (arXiv)

Ensuring robust safety measures across a wide range of scenarios is crucial for user-facing systems. While Large Language Models (LLMs) can generate valuable data for safety measures, they often exhibit distributional biases, focusing on common scenarios and neglecting rare but critical cases. This can undermine the effectiveness of safety protocols developed using such data. To address this, we propose a novel framework that integrates active learning with clustering to guide LLM generation, enhancing their representativeness and robustness in safety scenarios. We demonstrate the effectiveness of our approach by constructing a dataset of 5.4K potential safety violations through an iterative process involving LLM generation and an active learner model's feedback. Our results show that the proposed framework produces a more representative set of safety scenarios without requiring prior knowledge of the underlying data distribution. Additionally, data acquired through our method improves the accuracy and F1 score of both the active learner model as well models outside the scope of active learning process, highlighting its broad applicability.