Jialin Chen

I am a computer science Ph.D. student at Yale University advised by Professor Rex Ying. Before that, I obtained my B.S. degree at Shanghai Jiao Tong University (SJTU) in 2022. I majored in mathematics and applied mathematics (Zhiyuan Honors Program).

I am generally interested in Large Language Models (LLMs) with a focus on enabling reliable reasoning over complex, structured, and dynamic data, including:

Developing scalable foundation models with inductive biases for graph, relational, and structured data, moving beyond sequence-only architectures
Designing efficient structured Retrieval-Augmented Generation (RAG) and tool-integrated reasoning systems for multi-hop and decision-critical tasks
Exploring post-training strategies to improve faithfulness, calibration, and robustness
Advancing LLM capabilities in temporal reasoning and time-series understanding, bridging numerical dynamics with semantic modeling
Improving model trustworthiness, interpretability, and verification, particularly in high-stakes, real-world deployment scenarios

If you share similar interests or have any job opportunities in related areas, feel free to reach out.

Selected Publications

Interpreting Structural Models — Identifying Causal Structure

How to make structured models understandable and auditable: this theme studies mechanisms for identifying salient substructures, causal signals, and faithful explanations behind model decisions.

Theme 1

NeurIPS 2023	TempME: Towards the Explainability of Temporal Graph Neural Networks via Motif Discovery Jialin Chen, Rex Ying NeurIPS 2023 /
NeurIPS 2023	D4Explainer: In-distribution Explanations of Graph Neural Network via Discrete Denoising Diffusion Jialin Chen, Shirley Wu, Abhijit Gupta, Rex Ying NeurIPS 2023 /
IEEE	Generative Explanations for Graph Neural Network: Methods and Evaluations Jialin Chen, Kenza Amara, Junchi Yu, Rex Ying IEEE /

Reasoning over Structured Data — Aligning Structure with Language Models

How to align relational structure with language reasoning: this theme develops graph-native representations and retrieval pipelines that inject structural priors into LLMs for robust multi-hop reasoning.

Theme 2

WWW 2026	Towards A Universal Graph Structural Encoder Jialin Chen, Haolan Zuo, Haoyu Peter Wang, Siqi Miao, Pan Li, Rex Ying WWW 2026 /
EMNLP 2025	GRIL: Knowledge Graph Retrieval-Integrated Learning with Large Language Models Jialin Chen, H. Zhang, S. Yun, A. Mottini, R. Ying, X. Song, V. N. Ioannidis, Z. Li, Q. Cui EMNLP 2025 /
KDD 2025	LitFM: A Retrieval Augmented Structure-aware Foundation Model For Citation Graphs Jiasheng Zhang, Jialin Chen, Ali Maatouk, Ngoc Bui, Qianqian Xie, Leandros Tassiulas, Jie Shao, Hua Xu, Rex Ying KDD 2025 /
NeurIPS 2024 DB track	DTGB: A Comprehensive Benchmark for Dynamic Text-Attributed Graphs Jiasheng Zhang, Jialin Chen, Menglin Yang, Aosong Feng, Shuang Liang, Jie Shao, Rex Ying NeurIPS 2024 DB track /
COLING 2022	Modeling Hierarchical Reasoning Chains by Linking Discourse Units and Key Phrases for Reading Comprehension Jialin Chen, Zhuosheng Zhang, Hai Zhao COLING 2022 /

Temporal Reasoning — From Forecasting to Grounded Decision-Making

How to reason over evolving signals in context: this theme connects temporal dynamics with semantic grounding to support forecasting, retrieval, and decision-oriented reasoning in real-world settings.

Theme 3

NeurIPS 2025	TRACE: Grounding Time Series in Context for Multimodal Embedding and Retrieval Jialin Chen, Ziyu Zhao, Gaukhar Nurbek, Aosong Feng, Ali Maatouk, Leandros Tassiulas, Yifeng Gao, Rex Ying NeurIPS 2025 /
Preprint	MTBench: A Multimodal Time Series Benchmark for Temporal Reasoning and Question Answering Jialin Chen, Aosong Feng, Ziyu Zhao, Juan Garza, Gaukhar Nurbek, Cheng Qin, Ali Maatouk, Leandros Tassiulas, Yifeng Gao, Rex Ying Preprint /
NeurIPS 2024	From Similarity to Superiority: Channel Clustering for Time Series Forecasting Jialin Chen, Jan Eric Lenssen, Aosong Feng, Weihua Hu, Matthias Fey, Leandros Tassiulas, Jure Leskovec, Rex Ying NeurIPS 2024 /

Experience

Student Researcher at Google
2025.06 - 2026.01 | New York, NY
Designed a graph-structured LLM framework for scalable multi-table reasoning.
Introduced typed tool execution with a structure-aware post-training pipeline to improve reasoning robustness and faithfulness, with gains in accuracy, token efficiency, and out-of-schema generalization.

Applied Scientist Intern at Amazon
2024.05 - 2024.12 | Seattle, WA
Built an end-to-end graph retriever + LLM system for multi-hop reasoning on large knowledge graphs.
Improved retrieval scalability, precision, and LLM inference efficiency, enabling smaller models to match or outperform proprietary LLMs.

Machine Learning Intern at Kumo.AI
2023.05 - 2023.08 | Mountain View, CA
Improved time series forecasting by balancing per-channel specificity and cross-channel interactions.
Plugged into diverse forecasting backbones to boost accuracy, enable zero-shot generalization, and increase model interpretability.

Teaching

Teaching Assistant, 2023 Fall, Deep Learning on Graph-Structured Data (CPSC 483) Course Website
Teaching Assistant, 2023 Spring, Trustworthy Deep Learning (CPSC 680) Course Website

Academic Services

Conference Reviewer: NeurIPS, ICML, ICLR, KDD, AAAI, LoG
Journal Reviewer: IEEE Transactions on Artificial Intelligence, Journal of Biomedical Informatics
Organizer: NEGEL workshop at TheWebConf 2025 Website