Emergent Self-Awareness in Continual Robot Learning
Discover how continual learning fosters emergent self-awareness in robots by developing stable cognitive networks across tasks.
Boost Deepfake Audio Detection with Neuron Dropin & Plasticity
Improve deepfake audio detection accuracy and efficiency using neuron-level dropin and neuroplasticity mechanisms that reduce error rates and computation.
GameplayQA: Benchmarking Multi-Video 3D Agent Understanding
GameplayQA benchmarks AI decision-making in 3D multi-video environments, enhancing autonomous agents' perception and reasoning abilities.
Enhance Document Parsing with Coarse-to-Fine Visual Processing
Improve document parsing accuracy and speed using PaddleOCR-VL's coarse-to-fine visual processing to reduce redundancy and boost efficiency.
LLM-Guided Reward Design for Multi-Agent Cooperation
Discover how large language models optimize reward design to boost coordination in cooperative multi-agent reinforcement learning tasks.