15 papers · 1 filter
Routing Is Least Learnable Where It Is Most Valuable: Bounds on Representation Routing for Web Agents
Jiaming Wei, Zekun Wu, Adriano Koshiyama +1
Web agents observe a browser through text, pixels, or both, and the choice is usually fixed once for all tasks. We measure six observation modes across eight site-model combination…
Tool Calling is Linearly Readable and Steerable in Language Models
Zekun Wu, Ze Wang, Seonglae Cho +4
When a tool-calling agent picks the wrong tool, the failure is invisible until execution: the email gets sent, the meeting gets missed. As agents take on consequential actions, one…
Sell Me This Stock: Unsafe Recommendation Drift in LLM Agents
Zekun Wu, Adriano Koshiyama, Sahan Bulathwela +1
People increasingly use LLM agents for multi-turn financial recommendations, where the agent pulls market data through tools and tracks user preferences across turns. When tool out…
Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training
Figarri Keisha, Zekun Wu, Ze Wang +2
Large language models increasingly rely on synthetic data due to human-written content scarcity, yet recursive training on model-generated outputs leads to model collapse, a degene…
Personality as a Probe for LLM Evaluation: Method Trade-offs and Downstream Effects
Gunmay Handa, Zekun Wu, Adriano Koshiyama +1
Personality manipulation in large language models (LLMs) is increasingly applied in customer service and agentic scenarios, yet its mechanisms and trade-offs remain unclear. We pre…
CorrSteer: Generation-Time LLM Steering via Correlated Sparse Autoencoder Features
Seonglae Cho, Zekun Wu, Adriano Koshiyama
Sparse Autoencoders (SAEs) can extract interpretable features from large language models (LLMs) without supervision. However, their effectiveness in downstream steering tasks is li…