DAST: Difficulty-Aware Self-Training on Large Language Models

ACL ARR 2025 February Submission6865 Authors

Abstract: Present Large Language Models (LLM) self-training methods always under-sample on challenging queries, leading to inadequate learning on difficult problems which limits LLMs’ ability. Therefore, this work proposes a difficulty-aware self-training (DAST) framework that focuses on improving both the quantity and quality of self-generated responses on challenging queries during self-training. DAST is specified in three components: 1) sampling-based difficulty level estimation, 2) difficulty-aware data augmentation, and 3) the self-training algorithm using SFT and DPO respectively. Experiments on mathematical tasks demonstrate the effectiveness and generalization of DAST, highlighting the critical role of difficulty-aware strategies in advancing LLM self-training.

Paper Type: Short

Research Area: Generation

Research Area Keywords: Generation

Contribution Types: NLP engineering experiment

Languages Studied: English

Submission Number: 6865