From the 1 of 9 linked papers with an AI index.
9 papers
CoGate: Confidence-Gated Co-Decoding for Secure Code Generation
Minghao Hu, Lannan Luo, Allen Roush +1
The paper introduces CoGate, a method that uses the confidence of a security expert model to gate its influence during co-decoding for generating more secure code with large langua…
Refusal-Gated Decoding: Preserving Refusal Behavior Under High-Temperature Sampling
Phillip Howard, Xin Su, Allen Roush +2
High-temperature sampling is one of the primary mechanisms for increasing diversity in LLMs. Recent advances in truncation-based sampling techniques have helped mitigate drawbacks…
Mirage Probes: How Vision Models Fake Visual Understanding
Daniel Ben-Levi, Judah Goldfeder, Weiliang Zhao +5
Vision-language models (VLMs) can answer image-based questions confidently, and often correctly, even when no image is provided. This mirage behavior inflates benchmark scores with…
Measure what Matters: Psychometric Evaluation of AI with Situational Judgment Tests
Alexandra Yost, Shreyans Jain, Shivam Raval +6
Persona conditioning is widely used to steer large language model (LLM) behavior, but it is unclear whether it induces stable behavioral structure or superficial variation. We prop…
Do Multi-Agents Dream of Electric Screens? Achieving Perfect Accuracy on AndroidWorld Through Task Decomposition
Pierre-Louis Favreau, Jean-Pierre Lo, Clement Guiguet +5
We present Minitap, a multi-agent system that achieves 100% success on the AndroidWorld benchmark, the first to fully solve all 116 tasks and surpassing human performance (80%). We…
A superpersuasive autonomous policy debating system
Allen Roush, Devin Gonier, John Hines +4
The capacity for highly complex, evidence-based, and strategically adaptive persuasion remains a formidable great challenge for artificial intelligence. Previous work, like IBM Pro…