Papers by Xufeng Duan
Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models (2026.findings-acl)
Copied to clipboard
Hengyuan Zhang, Zhihao Zhang, Ercong Nie, Mingyang Wang, Zunhai Su, Yiwei Wang, Qianli Wang, Shuzhou Yuan, Xufeng Duan, Qibo Xue, Zeping Yu, Chenming Shang, Xiao Liang, Jing Xiong, Hui Shen, Chaofan Tao, Zhengwu Liu, Senjie Jin, Zhiheng Xi, Dongdong Zhang, Sophia Ananiadou, Tao Gui, Ruobing Xie, Hayden Kwok-Hay So, Hinrich Schuetze, Xuanjing Huang, Qi Zhang, Ngai Wong
| Challenge: | Existing literature on mechanistic interpretation (MI) treats it as an observational science, leaving practical applications underexplored. |
| Approach: | They propose a survey structured around the pipeline to identify and improve MI models. |
| Outcome: | The proposed framework enables tangible improvements in Alignment, Capability, and Efficiency. |
Large Language Models for Automated Literature Review: An Evaluation of Reference Generation, Abstract Writing, and Review Composition (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) are a promising solution to automate literature review writing tasks. |
| Approach: | They propose a framework to automatically evaluate the performance of large language models in three key tasks of literature review writing: reference generation, abstract writing, and literature review composition. |
| Outcome: | The proposed framework assesses the hallucination rates in generated references and measures the semantic coverage and factual consistency of the literature summaries and compositions against human-written counterparts. |
Unveiling Language Competence Neurons: A Psycholinguistic Approach to Model Interpretability (2025.coling-main)
Copied to clipboard
| Challenge: | a new study explores how large language models capture aspects of human linguis-tic ability . large language model performance is limited by the mechanisms behind their performance . |
| Approach: | They employ psycholinguistic paradigms to explore neuron-level representations in language models . they found that large language models exhibit human-like abilities in three tasks . |
| Outcome: | The proposed model shows human-like abilities in sound-shape association, gender association and implicit causality. |
Linguistic Minimal Pairs Elicit Linguistic Similarity in Large Language Models (2025.coling-main)
Copied to clipboard
| Challenge: | a new analysis leverages linguistic minimal pairs to probe the internal linguistic representations of Large Language Models (LLMs). |
| Approach: | They propose to use linguistic minimal pairs to probe the internal linguistic representations of Large Language Models (LLMs). |
| Outcome: | The proposed analysis reveals that linguistic similarity is significantly influenced by training data exposure, leading to higher cross-LLM agreement in higher-resource languages. |