最新論文

HiScene: Creating Hierarchical 3D Scenes with Isometric View Generation

Uni3C: Unifying Precisely 3D-Enhanced Camera and Human Motion Controls for Video Generation

Seeing from Another Perspective: Evaluating Multi-View Understanding in MLLMs

CCMNet: Leveraging Calibrated Color Correction Matrices for Cross-Camera Color Constancy

Describe Anything: Detailed Localized Image and Video Captioning

70% Size, 100% Accuracy: Lossless LLM Compression for Efficient GPU Inference via Dynamic-Length Float

CLASH: Evaluating Language Models on Judging High-Stakes Dilemmas from Multiple Perspectives

Analyzing LLMs' Knowledge Boundary Cognition Across Languages Through the Lens of Internal Representations

Learning Adaptive Parallel Reasoning with Language Models

PerceptionLM: Open-Access Data and Models for Detailed Visual Understanding