paper-with-me

Papers

Aligned but Blind: Alignment Increases Implicit Bias by Reducing Awareness of Race

2025-05-30 · Lihao Sun, Chengzhi Mao, Valentin Hofmann, Xuechunzi Bai

Although value-aligned language models (LMs) appear unbiased in explicit bias evaluations, they often exhibit stereotypes in implicit word association tasks, raising concerns about their fair usage. We investigate the mechanisms behind this discrepancy and find that alignment surprisingly amplifies implicit bias in model outputs. Specifically, we show that aligned LMs, unlike their unaligned counterparts, overlook racial concepts in early internal representations when the context is ambiguous. Not representing race likely fails to activate safety guardrails, leading to unintended biases. Inspired by this insight, we propose a new bias mitigation strategy that works by incentivizing the representation of racial concepts in the early model layers. In contrast to conventional mitigation methods of machine unlearning, our interventions find that steering the model to be more aware of racial concepts effectively mitigates implicit bias. Similar to race blindness in humans, ignoring racial nuances can inadvertently perpetuate subtle biases in LMs.

📄 PDF Abstract BibTeX arXiv:2506.00253

Code (1)

slhleosun/aligned-but-blind 공식 구현 pytorch

Tasks

Machine Unlearning

Methods 이 논문이 사용한 방법론

AWARE We propose to theoretically and empirically examine the effect of incorporating weighting schemes into walk-aggregating GNNs. To this end, we propose a simple, interpretable, and…

Similar Papers 제목 키워드 기반

Covert Bias: The Severity of Social Views' Unalignment in Language Models Towards Implicit and Explicit Opinion

2024-08-15 · Abeer Aldayel, Areej Alokaili, Rehab ALahmadi

While various approaches have recently been studied for bias identification, little is known about how implicit language that does not explicitly convey a viewpoint affects bias amplification in large language models. To…

Actions Speak Louder than Words: Agent Decisions Reveal Implicit Biases in Language Models

2025-01-29 · YuXuan Li, Hirokazu Shirado, Sauvik Das

While advances in fairness and alignment have helped mitigate overt biases exhibited by large language models (LLMs) when explicitly prompted, we hypothesize that these models may still exhibit implicit biases when simul…

Decision MakingFairness

MixRea: Benchmarking Explicit-Implicit Reasoning in Large Language Models

2026-05-19 · Yuanqing Cai, Ziyi Huang, Minhao Liu, Lixin Duan 외 arxiv

Large language models (LLMs) are increasingly integrated into high-stakes decision-making. Inspired by the theory of \emph{inattentional blindness} in human cognition, we investigate whether LLMs, trained on human-prefer…

Exploring the Effects of Alignment on Numerical Bias in Large Language Models

2026-01-23 · Ayako Sato, Hwichan Kim, Zhousi Chen, Masato Mita 외 arxiv

"LLM-as-a-judge," which utilizes large language models (LLMs) as evaluators, has proven effective in many evaluation tasks. However, evaluator LLMs exhibit numerical bias, a phenomenon where certain evaluation scores are…

The Root Shapes the Fruit: On the Persistence of Gender-Exclusive Harms in Aligned Language Models

2024-11-06 · Anaelia Ovalle, Krunoslav Lehman Pavasovic, Louis Martin, Luke Zettlemoyer 외

Natural-language assistants are designed to provide users with helpful responses while avoiding harmful outputs, largely achieved through alignment to human preferences. Yet there is limited understanding of whether alig…