Rethinking the Role of LLMs in Time Series Forecasting
Ningbo Key Laboratory of Spatial Intelligence（空间智能 Ningbo 关键实验室）；Digital Derivative, Institute of Digital Twin, Eastern Institute of Technology, Ningbo（数字衍生，数字孪生研究所，东部技术研究所，Ningbo）；Zhejiang University（浙江大学）
Large language models (LLMs) have been introduced to time series forecasting (TSF) to incorporate contextual knowledge beyond numerical signals. However, existing studies question whether LLMs provide genuine benefits, often reporting comparable performance without LLMs. We show that such conclusions stem from limited evaluation settings and do not hold at scale. We conduct a large-scale study of LLM-based TSF (LLM4TSF) across 8 billion observations, 17 forecasting scenarios, 4 horizons, multiple alignment strategies, and both in-domain and out-of-domain settings. Our results demonstrate that \emph{LLM4TS indeed improves forecasting performance}, with especially large gains in cross-domain generalization. Pre-alignment outperforming post-alignment in over 90\% of tasks. Both pretrained knowledge and model architecture of LLMs contribute and play complementary roles: pretraining is critical under distribution shifts, while architecture excels at modeling complex temporal dynamics. Moreover, under large-scale mixed distributions, a fully intact LLM becomes indispensable, as confirmed by token-level routing analysis and prompt-based improvements. Overall, Our findings overturn prior negative assessments, establish clear conditions under which LLMs are not only useful, and provide practical guidance for effective model design. We release our code at https://github.com/EIT-NLP/LLM4TSF.
Rethinking Semantic Parsing for Large Language Models: Enhancing LLM Performance with Semantic Hints
National Key Laboratory for Multimedia Information Processing, Peking University（北京大学多媒体信息处理国家重点实验室）；School of Software and Microelectronics, Peking University（北京大学软件与微电子学院）；Tsinghua University（清华大学）
Rethinking the Roles of Large Language Models in Chinese Grammatical Error Correction
Tsinghua Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院）；School of Computer Science and Engineering, Central South University（中南大学计算机科学与工程学院）；The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ)（广东省人工智能与数字经济实验室（深圳））；University of Illinois Chicago（芝加哥大学伊利诺伊分校）
All Circuits Lead to Rome: Rethinking Functional Anisotropy in Circuit and Sheaf Discovery for LLMs
Rutgers University（罗格斯大学）；Northwestern University（西北大学）；The Hong Kong University of Science and Technology (Guangzhou)（香港科学与技术大学（广州））；University of Toronto（多伦多大学）
Time Series Forecasting with LLMs: Understanding and Enhancing Model Capabilities
Rutgers University（罗格斯大学）；Shanghai Jiaotong University（上海交通大学）；University of Liverpool（利物浦大学）；New Jersey Institute of Technology（新泽西理工学院）；Xi’an Jiaotong-Liverpool University（西交利物浦大学）
Rethinking the Understanding Ability across LLMs through Mutual Information
Department of Industrial and Systems Engineering（工业与系统工程系）；University of Houston（休斯敦大学）；Bakar Computational Health Sciences Institute（巴卡计算健康科学研究所）；UCSF
Rethinking Personalization in Large Language Models at the Token Level
Institute for Artificial Intelligence, Peking University（人工智能研究院，北京大学）；State Key Lab of General Artificial Intelligence, School of Intelligence Science and Technology, Peking University（通用人工智能国家重点实验室，北京大学智能科学与技术学院）；School of Mathematical Sciences, Peking University（数学科学学院，北京大学）
Hypothetical Documents or Knowledge Leakage? Rethinking LLM-based Query Expansion
Department of Intelligent Semiconductors, Soongsil University（智能半导体系，顺斯尔大学）；School of AI Convergence, Soongsil University（人工智能融合学院，顺斯尔大学）；MAUM AI Inc.（MAUM人工智能公司）；Adobe Research, USA（Adobe美国研究实验室）
The Lottery LLM Hypothesis, Rethinking What Abilities Should LLM Compression Preserve?
The Hong Kong University of Science and Technology（香港科技大学）；The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；National University of Singapore（新加坡国立大学）
