paper-with-me

홈 › Papers

R1dacted: Investigating Local Censorship in DeepSeek's R1 Language Model

2025-05-19 · Ali Naseh, Harsh Chaudhari, Jaechul Roh, Mingshi Wu, Alina Oprea, Amir Houmansadr

DeepSeek recently released R1, a high-performing large language model (LLM) optimized for reasoning tasks. Despite its efficient training pipeline, R1 achieves competitive performance, even surpassing leading reasoning models like OpenAI's o1 on several benchmarks. However, emerging reports suggest that R1 refuses to answer certain prompts related to politically sensitive topics in China. While existing LLMs often implement safeguards to avoid generating harmful or offensive outputs, R1 represents a notable shift - exhibiting censorship-like behavior on politically charged queries. In this paper, we investigate this phenomenon by first introducing a large-scale set of heavily curated prompts that get censored by R1, covering a range of politically sensitive topics, but are not censored by other models. We then conduct a comprehensive analysis of R1's censorship patterns, examining their consistency, triggers, and variations across topics, prompt phrasing, and context. Beyond English-language queries, we explore censorship behavior in other languages. We also investigate the transferability of censorship to models distilled from the R1 language model. Finally, we propose techniques for bypassing or removing this censorship. Our findings reveal possible additional censorship integration likely shaped by design choices during training or alignment, raising concerns about transparency, bias, and governance in language model deployment.

📄 PDF Abstract BibTeX arXiv:2505.12625

Code (0)

등록된 구현이 없습니다.

Tasks

Language ModelingLanguage ModellingLarge Language Model

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

Information Suppression in Large Language Models: Auditing, Quantifying, and Characterizing Censorship in DeepSeek

2025-06-14 · Peiran Qiu, Siyi Zhou, Emilio Ferrara

This study examines information suppression mechanisms in DeepSeek, an open-source large language model (LLM) developed in China. We propose an auditing framework and use it to analyze the model's responses to 646 politi…

Language ModelingLanguage ModellingLarge Language Model

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 …

Discovering Forbidden Topics in Language Models

2025-05-23 · Can Rager, Chris Wendler, Rohit Gandikota, David Bau

Refusal discovery is the task of identifying the full set of topics that a language model refuses to discuss. We introduce this new problem setting and develop a refusal discovery method, LLM-crawler, that uses token pre…

Memorization

Towards Quantifying The Privacy Of Redacted Text

2024-10-10 · Vaibhav Gusain, Douglas Leith

In this paper we propose use of a k-anonymity-like approach for evaluating the privacy of redacted text. Given a piece of redacted text we use a state of the art transformer-based deep learning network to reconstruct the…

DiversitySentenceSentence Similarity

What Large Language Models Do Not Talk About: An Empirical Study of Moderation and Censorship Practices

2025-04-04 · Sander Noels, Guillaume Bied, Maarten Buyl, Alexander Rogiers 외

Large Language Models (LLMs) are increasingly deployed as gateways to information, yet their content moderation practices remain underexplored. This work investigates the extent to which LLMs refuse to answer or omit inf…

Diversity