提升模型电路可解释性，精准定位关键神经连接
BlackboxNLP-2025 MIB Shared Task: Improving Circuit Faithfulness via Better Edge Selection
- 通过自助法识别稳定贡献的神经边
- 用比例策略筛选强正向贡献边，兼顾准确与忠实
- 改用整数规划替代贪心算法，提升整体可靠性
机制可解释性中的主要挑战之一是电路发现，即确定模型中执行特定任务的部分。我们基于机制可解释性基准（MIB），提出三项关键改进：首先，使用自助法识别具有稳定归因分数的边；其次，引入基于比例的简单选择策略，优先保留高正向得分边，平衡性能与忠实度；第三，将标准贪心选择替换为整数线性规划公式。所提方法在多个MIB任务和模型上均生成更忠实的电路，优于以往方法。代码已公开于：https://github.com/technion-cs-nlp/MIB-Shared-Task。
原文摘要 · Abstract (English)
One of the main challenges in mechanistic interpretability is circuit discovery, determining which parts of a model perform a given task. We build on the Mechanistic Interpretability Benchmark (MIB) and propose three key improvements to circuit discovery. First, we use bootstrapping to identify edges with consistent attribution scores. Second, we introduce a simple ratio-based selection strategy to prioritize strong positive-scoring edges, balancing performance and faithfulness. Third, we replace the standard greedy selection with an integer linear programming formulation. Our methods yield more faithful circuits and outperform prior approaches across multiple MIB tasks and models. Our code is available at: https://github.com/technion-cs-nlp/MIB-Shared-Task.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。