paper-with-me

홈 › Papers

Steering the CensorShip: Uncovering Representation Vectors for LLM "Thought" Control

2025-04-23 · Hannah Cyberey, David Evans

Large language models (LLMs) have transformed the way we access information. These models are often tuned to refuse to comply with requests that are considered harmful and to produce responses that better align with the preferences of those who control the models. To understand how this "censorship" works. We use representation engineering techniques to study open-weights safety-tuned models. We present a method for finding a refusal--compliance vector that detects and controls the level of censorship in model outputs. We also analyze recent reasoning LLMs, distilled from DeepSeek-R1, and uncover an additional dimension of censorship through "thought suppression". We show a similar approach can be used to find a vector that suppresses the model's reasoning process, allowing us to remove censorship by applying the negative multiples of this vector. Our code is publicly available at: https://github.com/hannahxchen/llm-censorship-steering

📄 PDF Abstract BibTeX arXiv:2504.17130

Code (1)

hannahxchen/llm-censorship-steering 공식 구현 pytorch

Methods 이 논문이 사용한 방법론

ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…

Similar Papers 제목 키워드 기반

Uncovering Latent Chain of Thought Vectors in Language Models

2024-09-21 · Jason Zhang, Scott Viteri

As language models grow more influential and trusted in our society, our ability to reliably steer them toward favorable behaviors becomes increasingly paramount. For this, we investigate the technique of steering vector…

ARCGSM8KMMLU

Steering Risk Preferences in Large Language Models by Aligning Behavioral and Neural Representations

2025-05-16 · Jian-Qiao Zhu, Haijiang Yan, Thomas L. Griffiths

Changing the behavior of large language models (LLMs) can be as straightforward as editing the Transformer's residual streams using appropriately constructed "steering vectors." These modifications to internal neural act…

Nudging Hidden States: Training-Free Model Steering for Chain-of-Thought Reasoning in Large Audio-Language Models

2026-03-15 · Lok-Lam Ieong, Chia-Chien Chen, Chih-Kai Yang, Yu-Han Huang 외 arxiv

Chain-of-thought (CoT) prompting has been extended to large audio-language models (LALMs) to elicit reasoning, yet enhancing its effectiveness without training remains challenging. We study inference-time model steering …

Annotating the Chain-of-Thought: A Behavior-Labeled Dataset for AI Safety

2025-10-20 · Antonio-Gabriel Chacón Menke, Phan Xuan Tan, Eiji Kamioka arxiv

Recent work has highlighted the importance of monitoring chain-of-thought reasoning for AI safety; however, current approaches that analyze textual reasoning steps can miss subtle harmful patterns and may be circumvented…

Model Directions, Not Words: Mechanistic Topic Models Using Sparse Autoencoders

2025-07-31 · Carolina Zheng, Nicolas Beltran-Velez, Sweta Karlekar, Claudia Shi 외 arxiv

Traditional topic models are effective at uncovering latent themes in large text collections. However, due to their reliance on bag-of-words representations, they struggle to capture semantically abstract features. While…

Text GenerationTopic Models