Yifang (Mark) Chen

yc6990@nyu.edu
CV (PDF)

Hi! I am Yifang (Mark) Chen (陈义方), a senior at New York University Shanghai, majoring in Computer Science with a minor in Mathematics.

I am currently a Research Assistant at the Data Intelligence and Reasoning Lab at NYU Shanghai, advised by Prof. Qiaoyu Tan. I am also fortunate to be co-advised by Prof. Jinyang Li on my research project.

I am applying to PhD programs for Fall 2027. If my interests fit your group, feel free to email me.

Research Interests

Long-horizon LLM agents: Agents that plan, search, and use tools over many steps, and how reinforcement learning can teach them when to keep going, change course, or stop.
Efficient and trustworthy search agents: Knowing how much to search and when the gathered evidence is enough to answer, and evaluating cost and reliability alongside accuracy.
Multi-agent systems: How specialized agents divide work and coordinate, and when coordination actually helps, building on GraphMAS.
Graphs and structure for agents: Using graph structure as context, memory, or a tool that helps agents organize and reason over what they retrieve, extending my work on OMG-VLM.

Awards & Honors

2026 NYU Shanghai Recognition Award (Top 88)
2025 Baosteel Outstanding Student Scholarship (Top 2 across all majors and classes, ~0.25%)
2025 NYU Shanghai Recognition Award (Top 75, ~4%)

news

Sep 30, 2026	Our preprint GraphMAS, a benchmark of multi-agent coordination for graph learning, is now available on arXiv.
Aug 21, 2026	One Model, Many Graphs has been accepted to the EMNLP 2026 Main Conference. See you in Budapest.
Jul 22, 2026	Received the 2026 NYU Shanghai Recognition Award.
Jul 21, 2026	Our preprint One Model, Many Graphs is now available on arXiv, with code.
Nov 20, 2025	Received the 2025 Baosteel Outstanding Student Scholarship.

selected publications

Preprint

GraphMAS: A Systematic Benchmark of Multi-Agent Coordination for Graph Learning

Jiayi Yang, Yifang Chen, Yuanfu Sun, Xinyan Ge, and Qiaoyu Tan

Sep 2026

A benchmark of LLM-based multi-agent coordination for graph learning, showing that coordinating specialized graph-reasoning agents beats single-agent reasoning and that instance-adaptive specialist selection gives the best accuracy-efficiency trade-off.

Abs PDF

LLM-based multi-agent systems coordinate specialized reasoning through aggregation, interaction, and adaptive control, yet their potential for graph learning remains unexplored. Graph learning is a natural setting for such systems because useful evidence may arise from heterogeneous local, long-range, global structural, and semantic perspectives whose relevance varies across instances. We introduce GraphMAS, a systematic benchmark of multi-agent coordination for graph learning. GraphMAS builds a shared pool of graph reasoning specialists and organizes coordination along two dimensions, inter-agent interaction and runtime adaptivity, yielding four paradigms and seven representative coordination methods, evaluated under a unified protocol across seven text-attributed graphs, three domains, and two graph learning tasks. We find that heterogeneous graph perspectives are complementary, and that coordinating specialists improves over individual specialists and single-agent graph reasoning. However, richer inter-agent interaction does not reliably help, whereas instance-adaptive specialist selection yields the strongest accuracy-efficiency trade-off. We further show that coordination can be learned over a fixed specialist pool and transfers to held-out graphs.
EMNLP

One Model, Many Graphs: Learning over Attributed Graphs across Heterogeneous Modalities with Vision-Language Models

Jiayi Yang^*, Yifang Chen^*, Yuanfu Sun, Jiajin Liu, and Qiaoyu Tan

In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, 2026

A single vision-language model, equipped with structure-aware graph adapters, that learns over text-, image-, and multimodal-attributed graphs and generalizes to unseen graphs and modality schemas.

Abs PDF Code

Vision-language models (VLMs) provide a unified representation space for textual and visual information, yet their potential as general-purpose backbones for graph-structured data remains largely unexplored. In practice, attributed graphs exhibit substantial modality heterogeneity: some graphs contain only textual node attributes, others only visual attributes, while still others provide both. Existing graph learning approaches are typically designed for fixed modality schemas, requiring separate models for different settings. We present OMG-VLM (One Model, Many Graphs with Vision-Language Models), a unified framework that leverages a pretrained VLM as a shared backbone and introduces structure-aware graph adapters that integrate neighborhood information while remaining compatible with the VLM’s native embedding space. Extensive experiments across diverse domains show that OMG-VLM consistently outperforms state-of-the-art GNN- and LLM-based baselines on node classification and link prediction, while generalizing well to unseen graphs and varying modality schemas.