Image de couverture de livre

Eyes Wide Shut? Exploring the Visual Shortcomings of Multimodal LLMs

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Tong, Shengbang, Liu, Zhuang, Zhai, Yuexiang, Ma, Yi, LeCun, Yann, Xie, Saining
Format:	Preprint
Publié:	2024
Sujets:	Computer Vision and Pattern Recognition
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

MetaMorph: Multimodal Understanding and Generation via Instruction Tuning
par: Tong, Shengbang, et autres
Publié: (2024)

Scaling Language-Free Visual Representation Learning
par: Fan, David, et autres
Publié: (2025)

Fine-Tuning Large Vision-Language Models as Decision-Making Agents via Reinforcement Learning
par: Zhai, Yuexiang, et autres
Publié: (2024)

Scaling Text-to-Image Diffusion Transformers with Representation Autoencoders
par: Tong, Shengbang, et autres
Publié: (2026)

Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs
par: Tong, Shengbang, et autres
Publié: (2024)

Are They the Same? Exploring Visual Correspondence Shortcomings of Multimodal LLMs
par: Zhou, Yikang, et autres
Publié: (2025)

SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training
par: Chu, Tianzhe, et autres
Publié: (2025)

Diffusion Transformers with Representation Autoencoders
par: Zheng, Boyang, et autres
Publié: (2025)

LeJEPA: Provable and Scalable Self-Supervised Learning Without the Heuristics
par: Balestriero, Randall, et autres
Publié: (2025)

Learning by Reconstruction Produces Uninformative Features For Perception
par: Balestriero, Randall, et autres
Publié: (2024)

A hierarchical loss and its problems when classifying non-hierarchically
par: Wu, Cinna, et autres
Publié: (2017)

Video Representation Learning with Joint-Embedding Predictive Architectures
par: Drozdov, Katrina, et autres
Publié: (2024)

Cambrian-S: Towards Spatial Supersensing in Video
par: Yang, Shusheng, et autres
Publié: (2025)

The Entropy Enigma: Success and Failure of Entropy Minimization
par: Press, Ori, et autres
Publié: (2024)

Beyond Language Modeling: An Exploration of Multimodal Pretraining
par: Tong, Shengbang, et autres
Publié: (2026)

URLOST: Unsupervised Representation Learning without Stationarity or Topology
par: Yun, Zeyu, et autres
Publié: (2023)

Transformers without Normalization
par: Zhu, Jiachen, et autres
Publié: (2025)

RoboPEPP: Vision-Based Robot Pose and Joint Angle Estimation through Embedding Predictive Pre-Training
par: Goswami, Raktim Gautam, et autres
Publié: (2024)

Hierarchical World Models as Visual Whole-Body Humanoid Controllers
par: Hansen, Nicklas, et autres
Publié: (2024)

Gaussian Embeddings: How JEPAs Secretly Learn Your Data Density
par: Balestriero, Randall, et autres
Publié: (2025)

Learning and Leveraging World Models in Visual Representation Learning
par: Garrido, Quentin, et autres
Publié: (2024)

Exploiting Linear Structure Within Convolutional Networks for Efficient Evaluation
par: Denton, Remi, et autres
Publié: (2014)

Blockwise Self-Supervised Learning at Scale
par: Siddiqui, Shoaib Ahmed, et autres
Publié: (2023)

Asymmetric Idiosyncrasies in Multimodal Models
par: Tao, Muzi, et autres
Publié: (2026)

PooDLe: Pooled and dense self-supervised learning from naturalistic videos
par: Wang, Alex N., et autres
Publié: (2024)

Forgotten Polygons: Multimodal Large Language Models are Shape-Blind
par: Rudman, William, et autres
Publié: (2025)

Response Wide Shut: Surprising Observations in Basic Vision Language Model Capabilities
par: Chandhok, Shivam, et autres
Publié: (2024)

Learning Latent Action World Models In The Wild
par: Garrido, Quentin, et autres
Publié: (2026)

Rectified LpJEPA: Joint-Embedding Predictive Architectures with Sparse and Maximum-Entropy Representations
par: Kuang, Yilun, et autres
Publié: (2026)

Navigation World Models
par: Bar, Amir, et autres
Publié: (2024)

Variance-Covariance Regularization Improves Representation Learning
par: Zhu, Jiachen, et autres
Publié: (2023)

Rate-In: Information-Driven Adaptive Dropout Rates for Improved Inference-Time Uncertainty Estimation
par: Zeevi, Tal, et autres
Publié: (2024)

Revisiting Feature Prediction for Learning Visual Representations from Video
par: Bardes, Adrien, et autres
Publié: (2024)

Seeing from Another Perspective: Evaluating Multi-View Understanding in MLLMs
par: Yeh, Chun-Hsiao, et autres
Publié: (2025)

Improving Pre-trained Self-Supervised Embeddings Through Effective Entropy Maximization
par: Chakraborty, Deep, et autres
Publié: (2024)

Eyes Tell the Truth: GazeVal Highlights Shortcomings of Generative AI in Medical Imaging
par: Wong, David, et autres
Publié: (2025)

VisionFoundry: Teaching VLMs Visual Perception with Synthetic Images
par: Zhou, Guanyu, et autres
Publié: (2026)

Eyes Will Shut: A Vision-Based Next GPS Location Prediction Model by Reinforcement Learning from Visual Map Feed Back
par: Zhang, Ruixing, et autres
Publié: (2025)

White-Box Transformers via Sparse Rate Reduction: Compression Is All There Is?
par: Yu, Yaodong, et autres
Publié: (2023)

Flow Map Distillation Without Data
par: Tong, Shangyuan, et autres
Publié: (2025)