LGAIMay 24, 2025

Steering LLM Reasoning Through Bias-Only Adaptation

arXiv:2505.18706v313 citationsh-index: 7EMNLP
Originality Incremental advance
AI Analysis

This work reduces the cost and complexity of fine-tuning for reasoning tasks, though it is incremental as it builds on existing adapter and RL-tuning methods.

The paper tackled the problem of efficiently fine-tuning large language models for mathematical reasoning by training a single steering vector per layer with reinforcement learning while freezing base weights, achieving accuracy matching fully RL-tuned models with only about 0.0016% additional parameters on an 8 billion-parameter model.

We show that training a single $d$-dimensional steering vector per layer with reinforcement learning, while freezing all base weights, matches the accuracy of fully RL-tuned reasoning models on mathematical-reasoning tasks. On an 8 billion-parameter model this adds only $\approx 0.0016\%$ additional parameters and reproduces performance across a range of base models and mathematical-reasoning benchmarks. These results tighten the upper bound on the parameter budget required for high-level chain-of-thought reasoning, indicating that millions of adapter weights are unnecessary. The minimal trainable footprint reduces optimizer memory and inter-GPU communication, lowering the overall cost of fine-tuning. Moreover, a logit-lens analysis shows that the learned vectors amplify coherent token directions, providing clearer insight into the model's internal computations.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes