View PDF HTML (experimental)

Abstract:Time series analysis has recently been coupled with Large Language Models (LLMs) to leverage their reasoning and world knowledge capabilities, yet gains remain limited. We attribute this to a fundamental mismatch between existing task formulations and LLM strengths: most settings reduce time series understanding to curve-fitting systems, focusing on low-level prediction while ignoring the semantic, contextual, and reasoning-intensive nature of real-world temporal this http URL address these limitations, we introduce TSCognition, a multimodal benchmark for multi-dimensional time series reasoning. It collects real-world time series and textual information from 15 public sources and constructs approximately 41K QA samples around five cognitive reasoning tasks: Decoding, Grounding, Inferring, Extrapolating, and Acting. Building on this, we further propose TSAlign, a unified framework that encodes time series into compact patch-level representations and aligns them with semantic directions in the LLM embedding space via gated residual injection and multivariate this http URL show that TSAlign outperforms existing LLM, VLM, and time series QA baselines on TSCognition and the publicly available TimerBed benchmark while substantially reducing computational this http URL is available at: [this https URL](this https URL)

Submission history

From: Xin Qiu [view email]
[v1] Sat, 20 Jun 2026 16:08:44 UTC (1,899 KB)