AI Paper Atlas

Research map / Benchmarks / ETTh1

ETTh1 leaderboard

42 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on ETTh1. Best MSE: 0.3557 (TS-RAG (Chronos-Bolt), NeurIPS 2025).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

ETTh1 · MSE

37 papers, each with its best reported number for its own method; lower is better.

#MSE ↓MethodPaperTopic
10.3557TS-RAG (Chronos-Bolt)
Chronos-Bolt
TS-RAG: Retrieval-Augmented Generation based Time Series Foundation Models are Stronger Zero-Shot Forecaster
NeurIPS 2025 · Table: Long-term zero-shot for
Foundation models for time series
20.367RAFTRetrieval Augmented Time Series Forecasting
ICML 2025 · Appendix, Component Analysis
Causal and generative time series
30.389AMRC + SOFTS
SOFTS
Abstain Mask Retain Core: Time Series Prediction by Adaptive Masking Loss with Representation Consistency
NeurIPS 2025 · Ablation Study table
Long-term forecasting
40.396TimeBaseTimeBase: The Power of Minimalism in Efficient Long-term Time Series Forecasting
ICML 2025 · Table 2
Long-term forecasting
50.4DropPatch
PatchTST
Enhancing Masked Time-Series Modeling via Dropping Patches
AAAI 2025 · Table 1
Foundation models for time series
60.401SAMoVARLinear Transformers as VAR Models: Aligning Autoregressive Attention Mechanisms with Autoregressive Forecasting
ICML 2025 · Summary of Multivariate TSF Re
Long-term forecasting
70.401TFPSLearning Pattern-Specific Experts for Time Series Forecasting Under Patch-level Distribution Shift
NeurIPS 2025 · Table: Compared with foundatio
Long-term forecasting
80.405ChimeraChimera: Effectively Modeling Multivariate Time Series with 2-Dimensional State Space Models
NeurIPS 2024 · Table: Average Performance on
Long-term forecasting
90.407TVNetTVNet: A Novel Time Series Analysis Method Based on Dynamic Convolution and 3D-Variation
ICLR 2025 · Long-term forecasting table
Long-term forecasting
100.408Time-LLM
Llama-7B
Time-LLM: Time Series Forecasting by Reprogramming Large Language Models
ICLR 2024 · Table 1
Foundation models for time series
27 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

ETTh1 · MAE

10 papers, each with its best reported number for its own method; lower is better.

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.