Attractor FCM: A Novel Fuzzy Cognitive Map Model Integrating Physics Constraints
A latest arXiv paper proposes a novel Attractor Fuzzy Cognitive Map (Attractor FCM) built on gradient descent and Jacobi…
📂 Research
Latest articles in Research
Photonic AI Accelerators Move Toward Topology-Aware Very Large-Scale Architectures
A latest arXiv paper proposes a topology-aware design methodology for very large-scale photonic AI accelerators, aiming …
MAEO Framework: Bringing Multi-Objective Optimization to Large-Scale Engineering Applications
Researchers have proposed the Multiobjective Animorphic Ensemble Optimization (MAEO) framework, which employs an island-…
EdgeSpike: Spiking Neural Networks Enable Ultra-Low-Power Perception for Edge IoT
A research team has introduced the EdgeSpike framework, combining spiking neural networks with hardware-aware neural arc…
JI-ADF: A New Multimodal Framework for Skin Lesion Classification
A research team has proposed JI-ADF, a tri-modal deep learning framework that integrates dermoscopic images, clinical ph…
Research on Adversarial Transferability of Vision-Language Models in Autonomous Driving
A new study systematically analyzes the cross-architecture adversarial attack transferability of vision-language models …
VeraRetouch: A Fully Differentiable Lightweight Framework Revolutionizing AI Photo Retouching
A latest arXiv paper introduces the VeraRetouch framework, achieving fully differentiable multi-task reasoning-based pho…
COHERENCE Benchmark: Evaluating Image-Text Alignment in Multimodal Interleaved Contexts
A research team has proposed the COHERENCE benchmark, focusing on evaluating the fine-grained image-text alignment capab…
"Judge Before You Drive": VLA Critic-Centric Framework Revolutionizes Autonomous Driving Decision-Making
A new study proposes the Judge-Then-Drive framework, the first to introduce the "critic" capability of Vision-Language-A…
Breakthrough in Hyperspectral Image Classification: Global Spectral Supertoken Clustering Method
A latest arXiv paper proposes a hyperspectral image classification method based on global spectral supertoken clustering…
LLM-Driven Semantic Prototype Optimization: A New Paradigm of Iterative Definition Refinement for Zero-Shot Classification
A latest arXiv paper proposes an iterative semantic prototype optimization method powered by large language models. By a…
CasLayout: Cascaded Diffusion Framework Revolutionizes 3D Indoor Scene Generation
Researchers propose CasLayout, a cascaded diffusion framework that simulates human design cognition to generate 3D indoo…