Qi Liu
4 papers · Latest:
Robotics
RobotEQ: Transitioning from Passive Intelligence to Active Intelligence in Embodied AI
RobotEQ introduces the first benchmark for active intelligence, assessing if embodied AI can understand and adhere to social norms without explicit commands.
2605.06234
Computer VisionPerceptual Flow Network for Visually Grounded Reasoning
PFlowNet enhances visually grounded reasoning in LVLMs by decoupling perception and using RL with geometric shaping, achieving new SOTA results.
2605.02730
Natural Language ProcessingSeaEvo: Advancing Algorithm Discovery with Strategy Space Evolution
SeaEvo improves LLM-guided algorithm discovery by using explicit natural-language strategy descriptions to organize and guide evolutionary search.
2604.24372
Natural Language ProcessingStepPO: Step-Aligned Policy Optimization for Agentic Reinforcement Learning
StepPO introduces step-aligned policy optimization for Agentic RL, shifting from token-level to step-level MDPs to enhance LLM agent capabilities.
2604.18401
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.