Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks102 from public data
- Adversarial-Learned Loss for Domain Adaptation201
The confusion matrix is introduced, which is learned through an adversarial manner in ALDA, to reduce the gap and align the feature distributions and outperforms state-of-the-art approaches in four standard domain adaptation datasets.
- CenterCLIP185
A multi-segment token clustering algorithm to find the most representative tokens and drop the non-essential ones, which surpasses existing state-of-the-art by a large margin on typical text-video benchmarks, while reducing the training memory cost and accelerating the inference speed.
- Prompt as Triggers for Backdoor Attack: Examining the Vulnerability in Language Models158
This study proposes ProAttack, a novel and efficient method for performing clean-label backdoor attacks based on the prompt, which uses the prompt itself as a trigger, which does not require external triggers and ensures correct labeling of poisoned samples, improving the stealthy nature of the backdoor attack.
- CLIP4STR: A Simple Baseline for Scene Text Recognition With Pre-Trained Vision-Language Model74
This work transforms CLIP into a scene text reader and introduces CLIP4STR, a simple yet effective STR method built upon image and text encoders of CLIP, achieving state-of-the-art performance on 13 STR benchmarks.
- Test-Time Adaptation with CLIP Reward for Zero-Shot Generalization in Vision-Language Models62
This work proposes TTA with feedback to rectify the model output and prevent the model from becoming blindly confident, and builds different fully TTA pipelines with RLCF to improve the zero-shot generalization ability of various VLMs.
- Stock market trend prediction using dynamical Bayesian factor graph54
A new qualitative method using the concept of dynamical Bayesian factor graph for the prediction of stock market trend is developed and demonstrated by applying it in the analysis of market trends in the US stock market and Chinese stock market with good prediction results.
- AntiLeakBench: Preventing Data Contamination by Automatically Constructing Benchmarks with Updated Real-World Knowledge53
This paper proposes AntiLeak-Bench, an automated anti-leakage benchmarking framework that significantly reduces the cost of benchmark maintenance to accommodate emerging LLMs and highlights that data contamination likely exists before LLMs' cutoff time.
- 52
- Automated ICD coding for primary diagnosis via clinically interpretable machine learning51
The model has been demonstrated its effectiveness for automated primary diagnosis coding in the Chinese context and its results are interpretable and has the potential to assist clinical coders to improve coding efficiency and quality in Chinese inpatient settings.
- 39
- Toward Better Accuracy-Efficiency Trade-Offs: Divide and Co-Training37
It is argued that increasing the number of networks (ensemble) can achieve better accuracy-efficiency trade-offs than purely increasing the width and validated with 8 different neural architectures on common benchmarks through extensive experiments.
- 35
- ES-Net: Erasing Salient Parts to Learn More in Re-Identification29
A novel network, Erasing-Salient Net (ES-Net), to learn comprehensive features by erasing the salient areas in an image by using a trainable pooling layer P-pooling that generalizes global max and global average pooling.
- 29
- 24
- 23
- 21
- 21
- 21
- SCALoss: Side and Corner Aligned Loss for Bounding Box Regression15
Side Overlap (SO) loss is proposed by maximizing the side overlap of two bounding boxes, which puts more penalty for low overlapping bounding box cases, and the Corner Distance is added into the objective function to speed up the convergence.
- 14
- 14
- 13
- 9
- 9
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.