arXiv subjects

Zixin Sun

Publications and source records attributed to Zixin Sun.

3 recordsLinked to original sources

arXiv2026

Is Bitcoin A Hedge Against Central Banking? Evidence from AI-Driven Monetary Policy Expectations

This study investigates the transmission of monetary policy narratives to Bitcoin prices, distinguishing policy expectations from realized policy implementation. We introduce a weekly Monetary Policy Expectations (MPE) index derived from the Large Language Model (LLM)-based classification of 118,000+ market messages, providing a granular measure of hawkish and dovish monetary policy discourse. We demonstrate that changes in the MPE index provide evidence of significant linear predictive information for Bitcoin returns at short-to-medium horizons, with significant Granger causality at multiple lags. A Long Short-Term Memory (LSTM) framework combined with SHapley Additive exPlanations (SHAP) further identifies nonlinear and regime-dependent relationships between monetary-policy expectations and Bitcoin returns, indicating that Bitcoin functions as a sensitive barometer of central bank signaling. In particular, hawkish monetary-policy narratives are associated with negative price responses that are not accounted for by contemporaneous Federal Funds Rate adjustments. These findings highlight Bitcoin's structural sensitivity to global monetary discourse, establishing LLM-derived monetary-policy sentiment as a high-frequency measure of central-bank communication and as an informative leading macroeconomic indicator for the digital asset landscape.

econ.GN↗

arXiv2024

Speak It Out: Solving Symbol-Related Problems with Symbol-to-Language Conversion for Language Models

Symbols (or more broadly, non-natural language textual representations) such as numerical sequences, molecular formulas, and table delimiters widely exist, playing important roles in various tasks such as abstract reasoning, chemical property prediction, and table question answering. Despite the impressive natural language comprehension capabilities of large language models (LLMs), their reasoning abilities for symbols remain inadequate, which could attributed to the difference between symbol representations and general natural languages. We propose symbol-to-language (S2L), a tuning-free method that enables large language models to solve symbol-related problems with information expressed in natural language. Specifically, S2L first converts the symbols involved to language-based representations, which can be implemented by prompting LLMs or leveraging external tools, then these language-based representations are integrated into the original problem via direct substitution or concatenation, serving as useful input information for LLMs. We evaluate the S2L method using both API-based (GPT-4, ChatGPT) and open-source (OpenChat) models over eight symbol-related tasks, ranging from symbol-only abstract reasoning to sentiment analysis in social media. Experimental results show that S2L consistently leads to superior performance. For example, by employing S2L for GPT-4, there can be average significant improvements of +21.9% and +9.5% for subtasks in 1D-ARC and Dyck language, respectively. Codes and data are available at https://github.com/THUNLP-MT/symbol2language.

cs.CL↗

arXiv2021

AutoGL: A Library for Automated Graph Learning

Recent years have witnessed an upsurge in research interests and applications of machine learning on graphs. However, manually designing the optimal machine learning algorithms for different graph datasets and tasks is inflexible, labor-intensive, and requires expert knowledge, limiting its adaptivity and applicability. Automated machine learning (AutoML) on graphs, aiming to automatically design the optimal machine learning algorithm for a given graph dataset and task, has received considerable attention. However, none of the existing libraries can fully support AutoML on graphs. To fill this gap, we present Automated Graph Learning (AutoGL), the first dedicated library for automated machine learning on graphs. AutoGL is open-source, easy to use, and flexible to be extended. Specifically, we propose a three-layer architecture, consisting of backends to interface with devices, a complete automated graph learning pipeline, and supported graph applications. The automated machine learning pipeline further contains five functional modules: auto feature engineering, neural architecture search, hyper-parameter optimization, model training, and auto ensemble, covering the majority of existing AutoML methods on graphs. For each module, we provide numerous state-of-the-art methods and flexible base classes and APIs, which allow easy usage and customization. We further provide experimental results to showcase the usage of our AutoGL library. We also present AutoGL-light, a lightweight version of AutoGL to facilitate customizing pipelines and enriching applications, as well as benchmarks for graph neural architecture search. The codes of AutoGL are publicly available at https://github.com/THUMNLab/AutoGL.

cs.LG↗