8 papers
Societal Alignment Frameworks Can Improve LLM Alignment
Karolina StaÅczak, Nicholas Meade, Mehar Bhatia +14
Recent progress in large language models (LLMs) has focused on producing responses that meet human expectations and align with shared values - a process coined alignment. However,…
Toward Open Weight Models Without Risks: Separating Public and Private Capabilities in LLMs
Charbel El Feghali, Arkil Patel, Nicholas Meade +3
Open-weight Large Language Models (LLMs) enable scientific progress and broad deployment. However, they make it difficult to control access to sensitive capabilities. Current pract…
The Blind Spot of Agent Safety: How Benign User Instructions Expose Critical Vulnerabilities in Computer-Use Agents
Xuwei Ding, Skylar Zhai, Linxin Song +6
Computer-use agents (CUAs) can now autonomously complete complex tasks in real digital environments, but when misled, they can also be used to automate harmful actions programmatic…
DeepSeek-R1 Thoughtology: Let's think about LLM Reasoning
Sara Vera MarjanoviÄ, Arkil Patel, Vaibhav Adlakha +14
Large Reasoning Models like DeepSeek-R1 mark a fundamental shift in how LLMs approach complex problems. Instead of directly producing an answer for a given input, DeepSeek-R1 creat…
AgentRewardBench: Evaluating Automatic Evaluations of Web Agent Trajectories
Xing Han Lù, Amirhossein Kazemnejad, Nicholas Meade +7
Web agents enable users to perform tasks on web browsers through natural language interaction. Evaluating web agents trajectories is an important problem, since it helps us determi…
Investigating Adversarial Trigger Transfer in Large Language Models
Nicholas Meade, Arkil Patel, Siva Reddy
Recent work has developed optimization procedures to find token sequences, called adversarial triggers, which can elicit unsafe responses from aligned language models. These trigge…