通过消除偏置神经元缓解指令遵循语言模型的偏置
Mitigating Biases for Instruction-following Language Models via Bias Neurons Elimination
AI总结：
该研究定义并验证了偏置神经元的存在，提出 CRISPR 方法利用可解释性技术自动识别并消除指令遵循语言模型中的偏置神经元，在零样本设置下有效缓解偏置且不损失模型性能，仅消除至少三个神经元即可生效。
摘要：
指令遵循语言模型经常表现出不良的偏置。在语言模型的真实世界使用中，由于通过零样本示例提示使用了广泛的指令，这些不良偏置可能会被加速。为了解决这个问题，我们首先定义了显著影响偏置输出的偏置神经元，并经验性地证明了其存在性。此外，我们提出了一种新颖且实用的偏置缓解方法 CRISPR，以在指令遵循设置中消除语言模型的偏置神经元。CRISPR 自动确定偏置输出，并使用可解释性方法将影响偏置输出的神经元归类为偏置神经元。实验结果证明了我们的方法在零样本指令遵循设置下缓解偏置的有效性，且不会损失模型的任务性能和现有知识。实验结果揭示了我们方法的泛化能力，因为它在各种指令和数据集下表现出鲁棒性。令人惊讶的是，我们的方法仅通过消除少数神经元（至少三个）即可缓解语言模型中的偏置。
英文摘要：
Instruction-following language models often show undesirable biases. These undesirable biases may be accelerated in the real-world usage of language models, where a wide range of instructions is used through zero-shot example prompting. To solve this problem, we first define the bias neuron, which significantly affects biased outputs, and prove its existence empirically. Furthermore, we propose a novel and practical bias mitigation method, CRISPR, to eliminate bias neurons of language models in instruction-following settings. CRISPR automatically determines biased outputs and categorizes neurons that affect the biased outputs as bias neurons using an explainability method. Experimental results demonstrate the effectiveness of our method in mitigating biases under zero-shot instruction-following settings without losing the model's task performance and existing knowledge. The experimental results reveal the generalizability of our method as it shows robustness under various instructions and datasets. Surprisingly, our method can mitigate the bias in language models by eliminating only a few neurons (at least three).
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
