Zobrazeno 1 - 10
of 108
pro vyhledávání: '"Perozzi, Bryan"'
Autor:
Chen, Zhikai, Mao, Haitao, Liu, Jingzhe, Song, Yu, Li, Bingheng, Jin, Wei, Fatemi, Bahare, Tsitsulin, Anton, Perozzi, Bryan, Liu, Hui, Tang, Jiliang
Given the ubiquity of graph data and its applications in diverse domains, building a Graph Foundation Model (GFM) that can work well across different graphs and tasks with a unified backbone has recently garnered significant interests. A major obstac
Externí odkaz:
http://arxiv.org/abs/2406.10727
Autor:
Fatemi, Bahare, Kazemi, Mehran, Tsitsulin, Anton, Malkan, Karishma, Yim, Jinyeong, Palowitch, John, Seo, Sungyong, Halcrow, Jonathan, Perozzi, Bryan
Large language models (LLMs) have showcased remarkable reasoning capabilities, yet they remain susceptible to errors, particularly in temporal reasoning tasks involving complex temporal logic. Existing research has explored LLM performance on tempora
Externí odkaz:
http://arxiv.org/abs/2406.09170
Retrieval Augmented Generation (RAG) has greatly improved the performance of Large Language Model (LLM) responses by grounding generation with context from existing documents. These systems work well when documents are clearly relevant to a question
Externí odkaz:
http://arxiv.org/abs/2405.18414
Autor:
Sanford, Clayton, Fatemi, Bahare, Hall, Ethan, Tsitsulin, Anton, Kazemi, Mehran, Halcrow, Jonathan, Perozzi, Bryan, Mirrokni, Vahab
Which transformer scaling regimes are able to perfectly solve different classes of algorithmic problems? While tremendous empirical advances have been attained by transformer-based neural networks, a theoretical understanding of their algorithmic rea
Externí odkaz:
http://arxiv.org/abs/2405.18512
Autor:
Perozzi, Bryan, Fatemi, Bahare, Zelle, Dustin, Tsitsulin, Anton, Kazemi, Mehran, Al-Rfou, Rami, Halcrow, Jonathan
How can we best encode structured data into sequential form for use in large language models (LLMs)? In this work, we introduce a parameter-efficient method to explicitly represent structured data for LLMs. Our method, GraphToken, learns an encoding
Externí odkaz:
http://arxiv.org/abs/2402.05862
Autor:
Tsitsulin, Anton, Perozzi, Bryan
Graph learning methods help utilize implicit relationships among data items, thereby reducing training label requirements and improving task performance. However, determining the optimal graph structure for a particular learning task remains a challe
Externí odkaz:
http://arxiv.org/abs/2312.04762
Graphs are a powerful tool for representing and analyzing complex relationships in real-world applications such as social networks, recommender systems, and computational finance. Reasoning on graphs is essential for drawing inferences about the rela
Externí odkaz:
http://arxiv.org/abs/2310.04560
Autor:
Phothilimthana, Phitchaya Mangpo, Abu-El-Haija, Sami, Cao, Kaidi, Fatemi, Bahare, Burrows, Mike, Mendis, Charith, Perozzi, Bryan
Precise hardware performance models play a crucial role in code optimizations. They can assist compilers in making heuristic decisions or aid autotuners in identifying the optimal configuration for a given program. For example, the autotuner for XLA,
Externí odkaz:
http://arxiv.org/abs/2308.13490
Autor:
Fatemi, Bahare, Abu-El-Haija, Sami, Tsitsulin, Anton, Kazemi, Mehran, Zelle, Dustin, Bulut, Neslihan, Halcrow, Jonathan, Perozzi, Bryan
Graph neural networks (GNNs) demonstrate outstanding performance in a broad range of applications. While the majority of GNN applications assume that a graph structure is given, some recent methods substantially expanded the applicability of GNNs by
Externí odkaz:
http://arxiv.org/abs/2308.10737
Graphs are a representation of structured data that captures the relationships between sets of objects. With the ubiquity of available network data, there is increasing industrial and academic need to quickly analyze graphs with billions of nodes and
Externí odkaz:
http://arxiv.org/abs/2307.14490