A Self-Supervised Reinforcement Learning Approach for Fine-Tuning Large Language Models Using Cross-Attention Signals

We propose a novel reinforcement learning framework for post training large language models that does not rely on human in the loop feedback. Instead, our approach uses cross attention signals within the model itself to derive a self supervised reward, thereby guiding iterative fine tuning of the model policy. By analyzing how the model attends to the input prompt during generation, we construct measures of prompt coverage, focus, and coherence. We then use these measures to rank or score candidate responses, providing a reward signal that encourages the model to produce well aligned, on topic text. In empirical comparisons against standard policy gradient methods and RL fine tuning with synthetic preference models, our method shows significant gains in prompt relevance and consistency over a non RL baseline. While it does not yet match the performance of fully human supervised RLHF systems, it highlights an important direction for scaling alignment with minimal human labeling. We provide a detailed analysis, discuss potential limitations, and outline future work for combining cross-attention based signals with smaller amounts of human feedback.

@article{kiruluta2025_2502.10482,
  title={ A Self-Supervised Reinforcement Learning Approach for Fine-Tuning Large Language Models Using Cross-Attention Signals },
  author={ Andrew Kiruluta and Andreas Lemos and Priscilla Burity },
  journal={arXiv preprint arXiv:2502.10482},
  year={ 2025 }
}