Author name cluster

Zhuofeng Li

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

4 papers

1 author row

TMLR Journal 2026 Journal Article

StructEval: Benchmarking LLMs' Capabilities to Generate Structural Outputs

Jialin Yang
Dongfu Jiang
Tony He
Sherman Siu
Yuxuan Zhang
Disen Liao
Zhuofeng Li
Huaye Zeng

As Large Language Models (LLMs) become integral to software development workflows, their ability to generate structured outputs has become critically important. We introduce $\textbf{StructEval}$, a comprehensive benchmark for evaluating LLMs' capabilities in producing both non-renderable (JSON, YAML, CSV) and renderable (HTML, React, SVG) structured formats. Unlike prior benchmarks, StructEval systematically evaluates structural fidelity across diverse formats through two paradigms: $\textbf{1)}$ generation tasks, producing structured output from natural language prompts, and $\textbf{2)}$ conversion tasks, translating between structured formats. Our benchmark encompasses 18 formats and 44 types of task, with novel metrics for format adherence and structural correctness. Results reveal significant performance gaps—even state-of-the-art models like o1-mini achieve only $75.58$ average score, with open-source alternatives lagging approximately $10$ points behind. We find generation tasks more challenging than conversion tasks, and producing correct visual content more difficult than generating text-only structures.

PDF Details

TMLR Journal 2025 Journal Article

Avoiding Structural Pitfalls: Self-Supervised Low-Rank Feature Tuning for Graph Test-Time Adaptation

Haoxiang Zhang
Zhuofeng Li
Qiannan Zhang
Ziyi Kou
Juncheng Li
Shichao Pei

Pre-trained graph neural networks (GNNs) have demonstrated significant success in leveraging large-scale graph data to learn transferable representations. However, their performance often degrades under distribution shifts, particularly in real-world scenarios where test labels are unavailable. To address this challenge, we propose Graph Optimization via Augmented Transformations (GOAT), a novel self-supervised test-time tuning paradigm that adapts pre-trained GNNs to distribution-shifted test data by focusing exclusively on node feature transformations. By avoiding complex and often suboptimal graph structure transformations, GOAT overcomes the limitations of existing data-centric methods. To further address the issue of transformation collapse, where feature transformations converge to trivial solutions—such as when the test-time learned data-centric transformation degenerates into a constant or identity mapping across different inputs, we introduce a parameter-efficient low-rank adapter that generates diverse transformations tailored to individual input graphs. This design not only enhances adaptation performance but also improves interpretability by avoiding modifications to the graph structure. Through extensive experiments on six real-world datasets with diverse distribution shifts, we demonstrate that GOAT achieves consistent performance improvements across different pre-trained GNN backbones, outperforming state-of-the-art test-time adaptation methods.

PDF Details

YNIMG Journal 2025 Journal Article

Neural correlates of olfactory working memory in the human brain

Zhuofeng Li
Shu-bin Li
Shaozhen Tan
Lu-lu Liu
Chao Yan
Lai-quan Zou

Previous research has revealed that the insula, pallidum, thalamus, hippocampus, middle frontal gyrus, and supplementary motor area are activated during odor memory and that the performance of olfactory working memory is affected by the verbalization of odors. However, the neural mechanisms underlying olfactory working memory and the role of verbalization in olfactory working memory are not fully understood. Twenty-nine participants were enrolled in a study to complete olfactory and visual n-back tasks using high- and low-verbalizability stimuli while undergoing fMRI imaging. The behavioral results showed that the participants achieved greater accuracy in the visual rather than olfactory n-back task. We observed increased activation in the precentral gyrus, superior frontal gyrus, middle frontal gyrus, supplementary motor area, and inferior parietal gyrus during olfactory working memory. Interestingly, decreased activation was observed in the olfactory 2-back task versus the 0-back task. Moreover, the left angular gyrus and inferior parietal gyrus were more strongly activated during processing of olfactory working memory using high-verbalizability odors. In conclusion, olfactory working memory engages cross-modal regions to facilitate responses, is involved in the monitoring and manipulation of information during working memory, and boasts a unique activation pattern that is different from that of visual working memory. Semantic information supports the representation of odor information in the working memory system.

Details DOI

NeurIPS Conference 2024 Conference Paper

TEG-DB: A Comprehensive Dataset and Benchmark of Textual-Edge Graphs

Zhuofeng Li
Zixing Gou
Xiangnan Zhang
Zhongyuan Liu
Sirui Li
Yuntong Hu
Chen Ling
Zheng Zhang

Text-Attributed Graphs (TAGs) augment graph structures with natural language descriptions, facilitating detailed depictions of data and their interconnections across various real-world settings. However, existing TAG datasets predominantly feature textual information only at the nodes, with edges typically represented by mere binary or categorical attributes. This lack of rich textual edge annotations significantly limits the exploration of contextual relationships between entities, hindering deeper insights into graph-structured data. To address this gap, we introduce Textual-Edge Graphs Datasets and Benchmark (TEG-DB), a comprehensive and diverse collection of benchmark textual-edge datasets featuring rich textual descriptions on nodes and edges. The TEG-DB datasets are large-scale and encompass a wide range of domains, from citation networks to social networks. In addition, we conduct extensive benchmark experiments on TEG-DB to assess the extent to which current techniques, including pre-trained language models, graph neural networks, and their combinations, can utilize textual node and edge information. Our goal is to elicit advancements in textual-edge graph research, specifically in developing methodologies that exploit rich textual node and edge descriptions to enhance graph analysis and provide deeper insights into complex real-world networks. The entire TEG-DB project is publicly accessible as an open-source repository on Github, accessible at https: //github. com/Zhuofeng-Li/TEG-Benchmark.

PDF Details DOI