PowerBench: Measuring Language Model Bias in Power-shifting Requests
cs.LG
Submitted: 2026-10-01
Updated: 2026-10-01
Code: https://github.com/meflynn/troopdata
Project page: https://commoncrawl.github.io
Sources
- Poli-Bias: Understanding and Measuring Large Language Model Biases in International Political Conflicts
- TukaBench: A Culturally Grounded Jailbreak Benchmark for African Languages
- Inference-Time Reasoning Selectively Reduces Implicit Social Bias in Large Language Models
- It's the humans, not the data: Geopolitical bias in LLMs originates in post-training, amplified by the language of the prompt
- Is Power-Seeking AI an Existential Risk?
- Geopolitical alignment: Endorsement effects in large language models
- Multilingual Jailbreak Challenges in Large Language Models
- The Interlocutor Effect: Why LLMs Leak More Personal Data to Agents Than Humans
- Made-in China, Thinking in America:U.S. Values Persist in Chinese LLMs
- ROK-FORTRESS: Measuring the Effect of Geopolitical Transcreation for National Security and Public Safety
- Multilingual jailbreaking of LLMs using low-resource languages
- Adding Error Bars to Evals: A Statistical Approach to Language Model Evaluations
- The Illusion of Cross-Lingual Safety in Low-Resource Languages
- Agreement Metrics for LLM-as-Judge Evaluation: What to Report and Why
- What's in a Name? Auditing Large Language Models for Race and Gender Bias
- Open Challenges in Multi-Agent Security: Towards Secure Systems of Interacting AI Agents
- A StrongREJECT for Empty Jailbreaks
- Evaluating and Mitigating Discrimination in Language Model Decisions
- Measuring & Mitigating Over-Alignment for LLMs in Multilingual Criminal Law Courts
- Low-Resource Languages Jailbreak GPT-4
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks