Self-Consistency (SC) is an effective decoding strategy that improves the reasoning performance of Large Language Models (LLMs) by generating multiple chain-of-thought reasoning paths and selecting the final answer via majority voting. However, it suffers from substantial inference costs because it requires a large number of samples. To mitigate this issue, Difficulty-Adaptive Self-Consistency (DSC) was proposed to reduce unnecessary token usage for easy problems by adjusting the number of samples according to problem difficulty. However, DSC requires additional model calls and pre-sampling to estimate difficulty, and this process is repeated when applying to each dataset, leading to significant computational overhead. In this work, we propose Activation-Informed Difficulty-Aware Self-Consistency (ACTSC) to address these limitations. ACTSC leverages internal difficulty signals reflected in the feed-forward network neuron activations to construct a lightweight difficulty estimation probe, without any additional token generation or model calls. The probe dynamically adjusts the number of samples for SC and can be applied to new datasets without requiring pre-sampling for difficulty estimation. To validate its effectiveness, we conduct experiments on five benchmarks. Experimental results show that ACTSC effectively reduces inference costs while maintaining accuracy relative to existing methods.

翻译：自一致性（SC）是一种有效的解码策略，它通过生成多条思维链推理路径并采用多数投票机制选择最终答案，从而提升大语言模型（LLM）的推理性能。然而，该方法因需要大量采样而带来显著的推理开销。为缓解这一问题，难度自适应自一致性（DSC）被提出，其通过根据问题难度调整采样数量来减少对简单问题的冗余令牌使用。但DSC需要额外的模型调用和预采样来估计难度，且该过程在应用于每个数据集时需重复执行，导致较大的计算开销。本工作中，我们提出基于激活信息的难度感知自一致性（ACTSC）以解决上述局限。ACTSC利用前馈网络神经元激活中反映的内部难度信号，构建一个轻量级难度估计探针，无需任何额外的令牌生成或模型调用。该探针动态调整SC的采样数量，并可应用于新数据集而无需为难度估计进行预采样。为验证其有效性，我们在五个基准测试上进行了实验。实验结果表明，相对于现有方法，ACTSC在保持精度的同时有效降低了推理成本。