hiw3c 日报精选
arXiv cs.AI/CL/LG 网页快照

偏见头项衡量什么?头级偏见归因中的不可识别性

原文标题 · What Does a Bias Head Measure? Non-Identifiability in Head-Level Bias Attribution

作者: S. M Seefat Alam | 在大型语言模型中定位社会偏见至个体注意力头并通过消融实验消除该偏见的方法,通过其对公平度指标的影响得到验证,而非通过证明选择这些头的统计方法能衡量偏见的证据。我们在Qwen-2.5-7B-Instruct和Llama-3.1-8B-Instruct上对此进行了审查。头映射与标签盲响应长度映射相关,ρ = -0.950和-0.768。

来源:arXiv cs.AI/CL/LG · arxiv.org

正文为英文,可一键机器翻译(仅首次需要等待)

Computer Science > Computation and Language

Title: What Does a Bias Head Measure? Non-Identifiability in Head-Level Bias Attribution

Abstract: Methods that localise social bias in large language models to individual attention heads and remove it by ablation are validated by their effect on a fairness metric, never by evidence that the statistic selecting the heads measures bias. We audit this on Qwen-2.5-7B-Instruct and Llama-3.1-8B-Instruct. The head map correlates at $\rho = -0.950$ and $-0.768$ with a label-blind response-length map, apparently identifying a surface confound. An intervention that breaks the length-label coupling rejects this reading: with the prompt held fixed, response length explains 1% of the statistic's variance. What does explain it is prompt identity, at 64% of variance, in an ordering that is not monotonic in length. Within any fixed prompt the fairness label, response length and the presence of a protected-group token are collinear at $\phi = 0.95$, partly by definition of the labelling rule, so no observational analysis can attribute the map to any of them, and no method in this family performs an experiment that could. The substantive claims fail independently: bias selectivity is not late-layer-concentrated on either model under either reference reading, and no head set we tested reduces the unfairness rate when ablated. Its complement, scored on the opposite label, reliably destroys refusal behaviour ($p = 2.7\times10^{-5}$), demonstrating that the pipeline detects effects at this scale.

Submission history

Access Paper:

Additional Features

Current browse context:

References & Citations

BibTeX formatted citation

Bookmark

Bibliographic and Citation Tools

Code, Data and Media Associated with this Article

Demos

Recommenders and Search Tools

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .