Deep Learning
The emergence of clusters in self-attention dynamics
We show that particles, representing tokens, tend to cluster toward particular limiting objects as time tends to infinity.
Effective Targeted Attacks for Adversarial Self-Supervised Learning Minseon Kim
To achieve robustness in these frameworks, Kim et al.
QuantSR: Accurate Low-bit Quantization for Efficient Image Super-Resolution
In addition, we demonstrate the scheme's satisfactory architecture generality by providing QuantSR-C
Disambiguated Attention Embedding for Multi-Instance Partial-Label Learning Wei T ang
In many real-world tasks, the concerned objects can be represented as a multi-instance bag associated with a candidate label set, which consists of one ground-truth label and several false positive labels.
On-the-Fly Adapting Code Summarization on Trainable Cost-Effective Language Models
Deep learning models are emerging to summarize source code to comment for code documentation and program comprehension. We can achieve good performance by training the model on large training corpus.
R-divergence for Estimating Model-oriented Distribution Discrepancy
To address this question, we introduce R-divergence, designed to assess model-oriented distribution discrepancies.
Group Robust Classification Without Any Group Information
Firstly, these methods implicitly assume that all group combinations are represented during training. To illustrate this, we introduce a systematic generalization task on the MPI3D dataset and discover that current algorithms fail to improve the ERM baseline when combinations of observed attribute values are missing.