Search Results for author:
Found 19 papers, 16 papers with code
LoRA of Change: Learning to Generate LoRA for the Editing Instruction from A Single Before-After Image Pair
Furthermore, generalizable models for image editing with visual instructions typically require quad data, i. e., a before-after image pair, along with query and target images.
CARE Transformer: Mobile-Friendly Linear Visual Transformer via Decoupled Dual Interaction
In this paper, we propose a new de\textbf{C}oupled du\textbf{A}l-interactive linea\textbf{R} att\textbf{E}ntion (CARE) mechanism, revealing that features' decoupling and interaction can fully unleash the power of linear attention.
Robust Fine-tuning of Zero-shot Models via Variance Reduction
When fine-tuning zero-shot models like CLIP, our desideratum is for the fine-tuned model to excel in both in-distribution (ID) and out-of-distribution (OOD).
Typicalness-Aware Learning for Failure Detection
We observe that, with the cross-entropy loss, model predictions are optimized to align with the corresponding labels via increasing logit magnitude or refining logit direction.
Instruction Tuning-free Visual Token Complement for Multimodal LLMs
However, current research is inherently constrained by challenges such as the need for high-quality instruction pairs and the loss of visual information in image-to-text training objectives.
Doubly Abductive Counterfactual Inference for Text-based Image Editing
Through the lens of the formulation, we find that the crux of TBIE is that existing techniques hardly achieve a good trade-off between editability and fidelity, mainly due to the overfitting of the single-image fine-tuning.
Classes Are Not Equal: An Empirical Study on Image Recognition Fairness
Second, with the proposed concept of Model Prediction Bias, we investigate the origins of problematic representation during optimization.
Decoupled Kullback-Leibler Divergence Loss
In this paper, we delve deeper into the Kullback-Leibler (KL) Divergence loss and mathematically prove that it is equivalent to the Decoupled Kullback-Leibler (DKL) Divergence loss that consists of 1) a weighted Mean Square Error (wMSE) loss and 2) a Cross-Entropy loss incorporating soft labels.
Learning Context-aware Classifier for Semantic Segmentation
Semantic segmentation is still a challenging task for parsing diverse contexts in different scenes, thus the fixed classifier might not be able to well address varying feature distributions during testing.
Understanding Imbalanced Semantic Segmentation Through Neural Collapse
Based on our empirical and theoretical analysis, we point out that semantic segmentation naturally brings contextual correlation and imbalanced distribution among classes, which breaks the equiangular and maximally separated structure of neural collapse for both feature centers and classifiers.
Generalized Parametric Contrastive Learning
Based on theoretical analysis, we observe that supervised contrastive loss tends to bias high-frequency classes and thus increases the difficulty of imbalanced learning.
Ranked #7 on
Long-tail Learning
on iNaturalist 2018
Region Rebalance for Long-Tailed Semantic Segmentation
In this paper, we study the problem of class imbalance in semantic segmentation.
Ranked #22 on
Semantic Segmentation
on ADE20K
Rebalanced Siamese Contrastive Mining for Long-Tailed Recognition
Given the promising performance of contrastive learning, we propose Rebalanced Siamese Contrastive Mining (ResCom) to tackle imbalanced recognition.
Ranked #5 on
Long-tail Learning
on CIFAR-10-LT (ρ=10)
Adversarial Attacks on ML Defense Models Competition
Due to the vulnerability of deep neural networks (DNNs) to adversarial examples, a large number of defense techniques have been proposed to alleviate this problem in recent years.
Parametric Contrastive Learning
In this paper, we propose Parametric Contrastive Learning (PaCo) to tackle long-tailed recognition.
Ranked #14 on
Long-tail Learning
on iNaturalist 2018
Improving Calibration for Long-Tailed Recognition
Motivated by the fact that predicted probability distributions of classes are highly related to the numbers of class instances, we propose label-aware smoothing to deal with different degrees of over-confidence for classes and improve classifier learning.
Ranked #18 on
Long-tail Learning
on CIFAR-10-LT (ρ=10)
ResLT: Residual Learning for Long-tailed Recognition
From this perspective, the trivial solution utilizes different branches for the head, medium, and tail classes respectively, and then sums their outputs as the final results is not feasible.
Ranked #22 on
Long-tail Learning
on CIFAR-10-LT (ρ=10)
Learnable Boundary Guided Adversarial Training
Previous adversarial training raises model robustness under the compromise of accuracy on natural data.
Ranked #1 on
Adversarial Defense
on CIFAR-100
Fast and Practical Neural Architecture Search
In this paper, we propose a fast and practical neural architecture search (FPNAS) framework for automatic network design.
