PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
October 16, 20250 citationsOpen Access

Can Large Language Models Adequately Perform Symbolic Reasoning Over Time Series?

View Full Paper
ZLZewen LiuLanzhou UniversityJNJuntong NiEmory UniversityXTXianfeng TangAmazon (Germany)

Key Points

  • Empirical results highlight the effectiveness of large language models in symbolic reasoning tasks.
  • SymbolBench includes tasks like multivariate symbolic regression, Boolean network inference, and causal discovery.
  • Integrating large language models with genetic programming can enhance the symbolic reasoning process.
  • The study emphasizes the need for context alignment and domain knowledge in automated scientific discovery.

Abstract

Uncovering hidden symbolic laws from time series data, as an aspiration dating back to Kepler's discovery of planetary motion, remains a core challenge in scientific discovery and artificial intelligence. While Large Language Models show promise in structured reasoning tasks, their ability to infer interpretable, context-aligned symbolic structures from time series data is still underexplored. To systematically evaluate this capability, we introduce SymbolBench, a comprehensive benchmark designed to assess symbolic reasoning over real-world time series across three tasks: multivariate symbolic regression, Boolean network inference, and causal discovery. Unlike prior efforts limited to simple algebraic equations, SymbolBench spans a diverse set of symbolic forms with varying complexity. We further propose a unified framework that integrates LLMs with genetic programming to form a closed-loop symbolic reasoning system, where LLMs act both as predictors and evaluators. Our empirical results reveal key strengths and limitations of current models, highlighting the importance of combining domain knowledge, context alignment, and reasoning structure to improve LLMs in automated scientific discovery.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Liu et al. (2025) studied this question.

synapsesocial.com/papers/68f0f51d8dd8ea469b1d6fdbhttps://doi.org/10.48550/arxiv.2508.03963
Ask AI
Helpful
Bookmark
Share
View Full Paper