Yuxin Yang

Yuxin Yang

3rd year Ph.D. student @ USC

A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.

Pages

Posts

Future Blog Post

less than 1 minute read

Published:

This post will show up by default. To disable scheduling of future posts, edit config.yml and set future: false.

Blog Post number 4

less than 1 minute read

Published:

This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.

Blog Post number 3

less than 1 minute read

Published:

This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.

Blog Post number 2

less than 1 minute read

Published:

This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.

Blog Post number 1

less than 1 minute read

Published:

This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.

portfolio

publications

PA-GNN: Parameter-Adaptive Graph Neural Networks

Published in Workshop on Dynamic Neural Networks at ICML 2022, 2022

This paper designed a trainable node-specific aggregator that learns from node position and features to create an adaptive graph filter. Spotlight presentation at ICML 2022 Dynamic Neural Networks Workshop.

Recommended citation: **Yuxin Yang**, Yitao Liang, Muhan Zhang. "PA-GNN: Parameter-Adaptive Graph Neural Networks." Workshop on Dynamic Neural Networks in the 39th International Conference on Machine Learning (ICML-22-DyNN-Workshop), spotlight presentation, 2022. [Video]
Download Paper

RECIPE-TKG: From Sparse History to Structured Reasoning for LLM-based Temporal Knowledge Graph Completion

Published in Under Review, 2024

RECIPE-TKG is a lightweight and data-efficient framework that combines rule-based multi-hop retrieval, contrastive fine-tuning, and test-time semantic filtering for temporal knowledge graph completion using LLMs.

Recommended citation: Ömer Faruk Akgül, Feiyu Zhu, **Yuxin Yang**, Rajgopal Kannan, Viktor Prasanna. "RECIPE-TKG: From Sparse History to Structured Reasoning for LLM-based Temporal Knowledge Graph Completion." arXiv preprint arXiv:2505.17794, 2024.
Download Paper

Training Diverse Graph Experts for Ensembles: A Systematic Empirical Study

Published in KDD (Under Review), 2024

This paper presents a systematic empirical study on training diverse graph experts for ensemble methods, exploring different strategies to improve graph neural network performance through ensemble learning.

Recommended citation: **Yuxin Yang**, Hongkuan Zhou, Rajgopal Kannan, Viktor Prasanna. "Training Diverse Graph Experts for Ensembles: A Systematic Empirical Study." arXiv preprint arXiv:2510.18370, 2024.
Download Paper

Towards Ideal Temporal Graph Neural Networks: Evaluations and Conclusions after 10,000 GPU Hours

Published in VLDB 2025, 2024

This work proposed a practical approach to TGNN model comparison that compares models at the modular level with a standardized and optimized implementation framework. It further revealed interplay between modules and datasets.

Recommended citation: **Yuxin Yang**, Hongkuan Zhou, Rajgopal Kannan, Viktor Prasanna. "Towards Ideal Temporal Graph Neural Networks: Evaluations and Conclusions after 10,000 GPU Hours." Proceedings of the VLDB Endowment, Volume 18, Issue 4, 2025.
Download Paper

SPARC-RAG: Adaptive Sequential-Parallel Scaling with Context Management for Retrieval-Augmented Generation

Published in arXiv preprint, 2026

A multi-agent framework for improving retrieval-augmented generation systems on multi-hop questions through unified context management that coordinates sequential and parallel reasoning while avoiding context pollution.

Recommended citation: **Yuxin Yang**, Gangda Deng, Ömer Faruk Akgül, Nima Chitsazan, Yash Govilkar, Akasha Tigalappanavara, Shi-Xiong Zhang, Sambit Sahu, Viktor Prasanna. "SPARC-RAG: Adaptive Sequential-Parallel Scaling with Context Management for Retrieval-Augmented Generation." arXiv preprint arXiv:2602.00083, 2026.
Download Paper

SAGERec: Sampling and Gating for Enhanced Long-Tail Item Recommendations

Published in WSDM 2026, 2026

SAGERec introduces novel sampling and gating mechanisms to enhance recommendation performance for long-tail items, addressing the challenge of imbalanced item popularity distributions in recommendation systems.

Recommended citation: Abdulla Alshabanah, **Yuxin Yang**, Murali Annavaram. "SAGERec: Sampling and Gating for Enhanced Long-Tail Item Recommendations." Proceedings of the 19th ACM International Conference on Web Search and Data Mining (WSDM), 2026.

talks

teaching