Seong Joon Oh
- Papers
- 37
Cite
Notes
Only stored in your browser.
Authored papers
37Compositional Generalization Requires Linear, Orthogonal Representations in Vision Embedding Models
arXiv 2026
MEME: Multi-entity & Evolving Memory Evaluation
arXiv 2026
Sparse Autoencoders enable Robust and Interpretable Fine-tuning of CLIP models
arXiv 2026
When Do Diffusion Models learn to Generate Multiple Objects?
arXiv 2026
Enhancing Multi-Image Understanding through Delimiter Token Scaling
arXiv 2026
Half-Truths Break Similarity-Based Retrieval
arXiv 2026
It Takes Two: Complementary Self-Distillation for Contextual Integrity in LLMs
arXiv 2026
Privacy Collapse: Benign Fine-Tuning Can Break Contextual Privacy in Language Models
arXiv 2026
Leaky Thoughts: Large Reasoning Models Are Not Private Thinkers
arXiv 2025
Diffusion Classifiers Understand Compositionality, but Conditions Apply
arXiv 2025
Dr.LLM: Dynamic Layer Routing in LLMs
arXiv 2025
On the rankability of visual embeddings
arXiv 2025
CLIP Behaves like a Bag-of-Words Model Cross-modally but not Uni-modally
arXiv 2025
Are We Done with Object-Centric Learning?
arXiv 2025
Does Data Scaling Lead to Visual Compositional Generalization?
arXiv 2025
Calibrating Large Language Models Using Their Generations Only
arXiv 2024
Scaling Up Membership Inference: When and How Attacks Succeed on Large Language Models
arXiv 2024
TRAP: Targeted Random Adversarial Prompt Honeypot for Black-Box Identification
arXiv 2024
Do Deep Neural Network Solutions Form a Star Domain?
arXiv 2024
A Bayesian Approach To Analysing Training Data Attribution In Deep Learning
a-bayesian-approach-to-analysing-training
Probabilistic Contrastive Learning Recovers the Correct Aleatoric Uncertainty of Ambiguous Inputs
arXiv 2023
Neglected Free Lunch -- Learning Image Classifiers Using Annotation Byproducts
arXiv 2023
SelecMix: Debiased Learning by Contradicting-pair Sampling
arXiv 2022
Dataset Condensation via Efficient Synthetic-Data Parameterization
arXiv 2022
Weakly Supervised Semantic Segmentation using Out-of-Distribution Data
CVPR 2022 1
ECCV Caption: Correcting False Negatives by Collecting Machine-and-Human-verified Image-Caption Associations for MS-COCO
arXiv 2022
Re-labeling ImageNet: from Single to Multi-Labels, from Global to Localized Labels
CVPR 2021 1
Rethinking Spatial Dimensions of Vision Transformers
ICCV 2021 10
Probabilistic Embeddings for Cross-Modal Retrieval
CVPR 2021 1
Keep CALM and Improve Visual Feature Attribution
ICCV 2021 10
AdamP: Slowing Down the Slowdown for Momentum Optimizers on Scale-invariant Weights
adamp-slowing-down-the-slowdown-for-momentum
Evaluation for Weakly Supervised Object Localization: Protocol, Metrics, and Datasets
arXiv 2020
Reliable Fidelity and Diversity Metrics for Generative Models
ICML 2020 1
What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis
what-is-wrong-with-scene-text-recognition-1
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features
cutmix-regularization-strategy-to-train-1
Learning De-biased Representations with Biased Representations
ICML 2020 1
Modeling Uncertainty with Hedged Instance Embedding
arXiv 2018
Affiliations
Frequent co-authors
10from 37 papers