ZETTELKASTEN

Publications

Research publications, conference abstracts, and manuscripts. Within each year, entries appear newest first by venue or presentation date; standalone preprints use their first public release date. Undated manuscripts appear last. For the latest bibliographic record, see my Google Scholar profile.

2026

  1. LLMs Learn Better In-Context from Rules than from Examples

    Xiang Fu*, Seungmin Cho*, Yukyung Lee, Najoung Kim

    The 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP 2026)Main Conference data

  2. Coordinating LLMs via Debate Trees: Hierarchical Decomposition Improves Truthfulness

    Xiang Fu, Kevin Gold

    AAAI 2026 Bridge Program on Advancing LLM-Based Multi-Agent Collaboration (WMAC 2026)Poster Paper OpenReview poster

2025

  1. Who’s the Impostor? Multi-Agent Social Deduction for Evaluating LLM Social Reasoning

    Xiang Fu

    NeurIPS 2025 Workshop on Evaluating the Evolving LLM LifecycleWorkshop Paper code

  2. Humans have more sophisticated strategies for processing semantically anomalous word pairs than do Large Language Models

    Catherine L. Caldwell-Harris, Xiang Fu

    66th Annual Meeting of the Psychonomic SocietyConference Abstract and Oral Presentation

  3. M²IV: Towards Efficient and Fine-grained Multimodal In-Context Learning via Representation Engineering

    Yanshu Li*, Yi Cao*, Hongyang He, Qisen Cheng, Xiang Fu, Xi Xiao, Tianyang Wang, Ruixiang Tang

    Conference on Language Modeling (COLM 2025)Conference Paper arXiv

  4. Can an Easy-to-Hard Curriculum Make Reasoning Emerge in Small Language Models? Evidence from a Four-Stage Curriculum on GPT-2

    Xiang Fu

    arXiv:2505.11643 [cs.CL]Preprint poster

2022

  1. GomokuPro: An Implementation of Enhanced Machine Learning Algorithm Utilizing Convolutional Neural Network in Gomoku Strategy and Predictions Model

    Xiang Fu

    2022 7th International Conference on Intelligent Computing and Signal Processing (ICSP)Conference Paper, pp. 1671–1677 code

Manuscripts

* Equal contribution.

† Corresponding author.