AI Paper Atlas

Research map / Time series / Foundation models for time series

Adaptive Time Series Reasoning via Segment Selection

Shvat Messica, Jiawen Zhang, Kevin Li, Theodoros Tsiligkaridis, Marinka Zitnik

ICML 2026 · Time series › Foundation models for time series · approach: Time-series language reasoning · 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
add RL stage after SFT (ablation)ARTIST · Average over six datasetsaccuracy (%): 63.61 -> 69.26Improved
add adaptive segment selection controller (ablation)ARTIST · ECG-QA + RCW averageaccuracy (%): 9.30% average accuracy drop without controllerImproved
add joint reasoner training (unfrozen reasoner) in RL (ablation)ARTIST · ECG-QA + RCW averageaccuracy (%): 8.94% decrease with controller-only RLImproved
add reliability reward for controller (ablation)ARTIST · ECG-QA + RCW averageaccuracy (%): 73.41 -> 51.97 when removed (21.44% drop)Improved
add trajectory-level (hierarchical) controller objective instead of myopic (ablation)ARTIST · ECG-QA + RCW averageaccuracy (%): 12.28% drop with myopic objectiveImproved
add variance-guided sampling for reasoner group selection (ablation)ARTIST · ECG-QA + RCW averageaccuracy (%): 73.41 -> 70.44 when removed (2.97%)Improved
use VLM backbone with plotted time series instead of tokenized inputARTIST (SFT only) · Sleep-QAaccuracy (%): 28.13 -> 65.00Improved
use ARTIST segment selection instead of PixelReasoner on Qwen3-4BQwen3-4B · ECG-QAaccuracy (%): 60.15 -> 69.81Improved
use ARTIST instead of strongest baselineARTIST · Average over six datasetsaccuracy (%): +6.46Improved
add more of the time series (near-complete usage)ARTIST · Sleep-QA and TRQAaccuracy (%)Hurt
add RL on TimeMaster vs ARTIST on Sleep-QATimeMaster · Sleep-QAaccuracy (%): TimeMaster+RL 72.55 vs ARTIST 36.63Mixed
add statistics to promptGPT-5 · ETIaccuracy (%): 29.50 -> 63.54Improved

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