ZINA: Multimodal Fine-grained Hallucination Detection and Editing
arXiv:2506.13130v1 Announce Type: cross Abstract: Multimodal Large Language Models (MLLMs) often generate hallucinations, where the output deviates from the visual…
Autonomous Computer Vision Development with Agentic AI
arXiv:2506.11140v2 Announce Type: replace-cross Abstract: Agentic Artificial Intelligence (AI) systems leveraging Large Language Models (LLMs) exhibit significant potential for complex…
MAGNET: A Multi-agent Framework for Finding Audio-Visual Needles by Reasoning over Multi-Video Haystacks
arXiv:2506.07016v2 Announce Type: replace-cross Abstract: Large multimodal models (LMMs) have shown remarkable progress in audio-visual understanding, yet they struggle with…
A Synthetic Pseudo-Autoencoder Invites Examination of Tacit Assumptions in Neural Network Design
arXiv:2506.12076v1 Announce Type: cross Abstract: We present a handcrafted neural network that, without training, solves the seemingly difficult problem of…
A Survey of Task-Oriented Knowledge Graph Reasoning: Status, Applications, and Prospects
arXiv:2506.11012v1 Announce Type: new Abstract: Knowledge graphs (KGs) have emerged as a powerful paradigm for structuring and leveraging diverse real-world…
Accelerating Diffusion Large Language Models with SlowFast Sampling: The Three Golden Principles
arXiv:2506.10848v2 Announce Type: replace-cross Abstract: Diffusion-based language models (dLLMs) have emerged as a promising alternative to traditional autoregressive LLMs by…
FIMA-Q: Post-Training Quantization for Vision Transformers by Fisher Information Matrix Approximation
arXiv:2506.11543v1 Announce Type: cross Abstract: Post-training quantization (PTQ) has stood out as a cost-effective and promising model compression paradigm in…
Foundation Models in Autonomous Driving: A Survey on Scenario Generation and Scenario Analysis
arXiv:2506.11526v1 Announce Type: cross Abstract: For autonomous vehicles, safe navigation in complex environments depends on handling a broad range of…
PiPViT: Patch-based Visual Interpretable Prototypes for Retinal Image Analysis
arXiv:2506.10669v2 Announce Type: replace-cross Abstract: Background and Objective: Prototype-based methods improve interpretability by learning fine-grained part-prototypes; however, their visualization in…