AI.info

Volker Tresp

Explore Volker Tresp on AI.info.

• ReflectRL: Learning from Golden Negative Trajectories via Reflective-to-Direct Reasoning
• MADA-RL: Multi-Agent Debate-Aware Reinforcement Learning for Parameter-Efficient Reasoning in Compact Models
• WebArbiter: A Principle-Guided Reasoning Process Reward Model for Web Agents
• AUVIC: Adversarial Unlearning of Visual Concepts for Multi-modal Large Language Models
• Improving Perturbation-based Explanations by Understanding the Role of Uncertainty Calibration