AI Paper Atlas

Research map / Time series / Foundation models for time series

Time Series Reasoning via Process-Verifiable Thinking Data Synthesis and Scheduling for Tailored LLM Reasoning

Jiahui Zhou, Dan Li, Boxin Li, Xiao Zhang, Erli Meng, Lin Li, Zhuomin Chen, Jian Lou, See-Kiong Ng

ICML 2026 · Time series › Foundation models for time series · approach: LLM-based time series forecasting · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
train on TSRBench process-verifiable CoT with SFT+RL instead of base modelQwen2.5-3B-Instruct · TSRBench scenario-based overallaccuracy (%): 40.99 -> 82.86Improved
train with VeriTime instead of base modelQwen3-4B-Instruct · TSRBench scenario-based overallaccuracy (%): 75.48 -> 86.55Improved
use TSRBench CoT reasoning data instead of original concise ChatTS Q&A pairs for SFT (ablation)Qwen3-4B-Instruct · TSRBench inferential calculationaccuracy (%): 72.38 -> 77.14Improved
add GRPO RL stage after SFT (ablation)VeriTime on Qwen3-4B-Instruct · TimeSeriesExamaccuracy (%): 40.31 -> 47.27Improved
add GRPO RL stage after SFT (ablation)VeriTime on Qwen2.5-3B-Instruct · DROPaccuracy (%): 66.44 -> 82.55Improved
apply TS-tailored CoT with VeriTime to reduce reasoning lengthQwen3-4B-Instruct · 8 tasks inferenceaverage token count: 2192.46 -> 632.67 (anomaly detection)Improved
remove hard reward (final answer correctness) from RL (ablation)VeriTime on Qwen2.5-3B-Instruct · TSRBench inferential calculationaccuracy (%): 47.6 -> 68.6Improved
add critical pattern reward (ablation)VeriTime on Qwen2.5-3B-Instruct · TSRBench scenario-based tasksaccuracy (%): 18.94% average drop when removedImproved
add answer alignment and verification rewards (ablation)VeriTime on Qwen2.5-3B-Instruct · TSRBench inferential calculationaccuracy (%): 23.62% drop when removedImproved
add structural rewards (format and length) (ablation)VeriTime on Qwen2.5-3B-Instruct · TSRBench scenario-based tasksaccuracy (%)Improved
model-guided data scheduling (failed cases to RL, successful to SFT) instead of random split (ablation)VeriTime on Qwen3-4B-Instruct · TSRBench scenario-based overallaccuracy (%): 83.30 -> 86.55Improved
apply full RL on all scenario-based data instead of model-guided scheduling (ablation)VeriTime · TSRBench scenario-based tasksaccuracy (%)Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 3 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy