Transformation invariant learning and theoretical guarantees for the generalization of OOD
Learning with identical distributions of trains and trials has been extensively investigated in both practice and theory. However, much remains ...
Tag: Generalization
Add and adapt natural language cues for subsequent CLIP generalization
Large pre-trained vision and language models, such as CLIP, have shown promising generalization ability, but may struggle in specialized domains ...
KAUST researchers use Anderson exploit to maximize GPU efficiency with increased model accuracy and generalization
The scaling of ai means greater spending on infrastructure. Massive, multidisciplinary research puts economic pressure on institutions, as high-performance computing ...
On the limited generalization capacity of the implicit reward model induced by direct preference optimization.
Reinforcement learning from human feedback (RLHF) is an effective approach to align language models with human preferences. Fundamental to RLHF ...
Google DeepMind researchers propose human-centric alignment for vision models to boost AI generalization and interpretability
Deep learning has made significant advances in artificial intelligence, particularly in natural language processing and computer vision. However, even the ...
On the benefits of pixel-based hierarchical policies for task generalization
Reinforcement learning practitioners typically avoid hierarchical policies, especially in image-based observation spaces. Typically, the performance improvement on a single task ...
Towards Generalization on Graphs: From Invariance to Causality | by Qitian Wu | Jul, 2024
Graph machine learning remains a popular research direction, especially with the wave of AI4Science driving increasingly diverse applications of graph ...
Exploring sharpness-aware minimization (SAM): insights into label noise robustness and generalization
Recently, there has been increasing interest in improving the generalization of deep networks by regulating the sharpness of the loss ...
Overcoming the obstacles of fitting the vision-language model for the generalization of OOD
Existing models of vision and language exhibit strong generalization across a variety of visual domains and tasks. However, these models ...
ByteDance AI Research presents a reinforced fine-tuning (ReFT) method to improve the generalization of LLM learning for reasoning with mathematical problem solving as an example
An effective method to improve LLMs' reasoning skills is to employ supervised fine-tuning (SFT) with chain-of-thought (CoT) annotations. However, this ...
Recommended.
