Aware Fragment

"Aware Fragment" research focuses on improving the processing of large datasets by intelligently managing and utilizing smaller, meaningful segments or "fragments" of data. Current work explores methods to identify and prioritize relevant fragments based on their relationships to each other (e.g., in narratives or code) and their contribution to overall meaning, often employing transformer models and convolutional neural networks. This approach enhances efficiency and accuracy in various applications, including long-text understanding, drug discovery (via molecular fragment analysis), and image-text retrieval by improving the alignment of relevant features.

Papers

June 5, 2024

FragRel: Exploiting Fragment-level Relations in the External Memory of Large Language Models
Xihang Yue, Linchao Zhu, Yi Yang
Unstructured Text Hierarchical Memory Memory Capacity Aware Fragment

January 15, 2024

Empirical Evidence for the Fragment level Understanding on Drug Molecular Structure of LLMs
Xiuyuan Hu, Guoqing Liu, Yang Zhao, Hao Zhang
Language Model Medical LLM Molecular Design Chemical Space Empirical Evidence Drug Molecule Fuzzy String Matching Aware Fragment

November 3, 2023

A New Fine-grained Alignment Method for Image-text Matching
Yang Zhang
Fine Grained Image Text Retrieval Image Text Matching Fine Grained Cross Modal Alignment Aware Fragment

October 2, 2023

Drug Discovery with Dynamic Goal-aware Fragments
Seul Lee, Seanie Lee, Kenji Kawaguchi, Sung Ju Hwang
Drug Discovery Molecular Generative Aware Fragment Drug Discovery Task

April 14, 2023

A Byte Sequence is Worth an Image: CNN for File Fragment Classification Using Bit Shift and n-Gram Embeddings
Wenyang Liu, Yi Wang, Kejun Wu, Kim-Hui Yap, Lap-Pui Chau
CNN Network N Gram Sub Byte Bit Flip File Classification Aware Fragment Intra Byte

Aware Fragment

Papers

FragRel: Exploiting Fragment-level Relations in the External Memory of Large Language Models

Empirical Evidence for the Fragment level Understanding on Drug Molecular Structure of LLMs

A New Fine-grained Alignment Method for Image-text Matching

Drug Discovery with Dynamic Goal-aware Fragments

A Byte Sequence is Worth an Image: CNN for File Fragment Classification Using Bit Shift and n-Gram Embeddings