empirical studies
Empirical studies are systematic investigations that derive knowledge from experimentation and observation in AI. These studies help in understanding model performance, strengths, and limitations in diverse scenarios.
- A Beyond-Worst-Case Analysis of Greedy k-means++
- ACT as Human: Multimodal Large Language Model Data Annotation with Critical Thinking
- Co-Regularization Enhances Knowledge Transfer in High Dimensions
- Efficient Adaptive Experimentation with Noncompliance
- Feature Unlearning: Theoretical Foundations and Practical Applications with Shuffling
- Fourier Token Merging: Understanding and Capitalizing Frequency Domain for Efficient Image Generation
- Graph Data Selection for Domain Adaptation: A Model-Free Approach
- HypoBootstrap: A Bootstrapping Framework for Inductive Reasoning
- Is Limited Participant Diversity Impeding EEG-based Machine Learning?
- Less is More: Unlocking Specialization of Time Series Foundation Models via Structured Pruning
- MultiNet: Adaptive Multi-Viewed Subgraph Convolutional Networks for Graph Classification
- On the Value of Cross-Modal Misalignment in Multimodal Representation Learning
- PDEfuncta: Spectrally-Aware Neural Representation for PDE Solution Modeling
- RoMA: Scaling up Mamba-based Foundation Models for Remote Sensing
- Self-Verification Provably Prevents Model Collapse in Recursive Synthetic Training
- Streaming Federated Learning with Markovian Data
- TS-RAG: Retrieval-Augmented Generation based Time Series Foundation Models are Stronger Zero-Shot Forecaster
- Towards Implicit Aggregation: Robust Image Representation for Place Recognition in the Transformer Era
- d1: Scaling Reasoning in Diffusion Large Language Models via Reinforcement Learning