# TimeHF: Billion-Scale Time Series Models Guided by Human Feedback

arXiv: https://arxiv.org/abs/2501.15942 · HTML: https://mlsanity.com/paper/2501.15942/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Placed in Reward modeling and RLHF, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 17 of its 22 references that are in the map.
- First posted on arXiv 27 Jan 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Reward modeling and RLHF](https://mlsanity.com/alignment/reward-modeling-and-rlhf/)

## Builds on (most cited first, 17 of 17)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis](https://mlsanity.com/paper/2210.02186/) (2022)
- [Chronos: Learning the Language of Time Series](https://mlsanity.com/paper/2403.07815/) (2024)
- [Time-LLM: Time Series Forecasting by Reprogramming Large Language Models](https://mlsanity.com/paper/2310.01728/) (2023)
- [A decoder-only foundation model for time-series forecasting](https://mlsanity.com/paper/2310.10688/) (2023)
- [One Fits All:Power General Time Series Analysis by Pretrained LM](https://mlsanity.com/paper/2302.11939/) (2023)
- [Unified Training of Universal Time Series Forecasting Transformers](https://mlsanity.com/paper/2402.02592/) (2024)
- [Monash Time Series Forecasting Archive](https://mlsanity.com/paper/2105.06643/) (2021)
- [Timer: Generative Pre-trained Transformers Are Large Time Series Models](https://mlsanity.com/paper/2402.02368/) (2024)
- [Revisiting Neural Scaling Laws in Language and Vision](https://mlsanity.com/paper/2209.06640/) (2022)
- [VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasters](https://mlsanity.com/paper/2408.17253/) (2024)
- [TimeRAG: BOOSTING LLM Time Series Forecasting via Retrieval-Augmented Generation](https://mlsanity.com/paper/2412.16643/) (2024)

## Built on by (most cited first, 10 of 10)

- [Empowering Time Series Analysis with Synthetic Data: A Survey and Outlook in the Era of Foundation Models](https://mlsanity.com/paper/2503.11411/) (2025)
- [Text Reinforcement for Multimodal Time Series Forecasting](https://mlsanity.com/paper/2509.00687/) (2025)
- [TimeRFT: Stimulating Generalizable Time Series Forecasting for TSFMs via Reinforcement Finetuning](https://mlsanity.com/paper/2605.00015/) (2026)
- [TF-CoDiT: Conditional Time Series Synthesis with Diffusion Transformers for Treasury Futures](https://mlsanity.com/paper/2601.11880/) (2026)
- [Ground-Truth Neighborhood Regularization for Reinforcement Learning Post-Training of Time Series Foundation Models](https://mlsanity.com/paper/2608.08010/) (2026)
- [Post-Training Corrections for Improved Time-Series Forecasting](https://mlsanity.com/paper/2505.15354/) (2025)
- [Forecast-Then-Optimize Deep Learning Methods](https://mlsanity.com/paper/2506.13036/) (2025)
- [There is No "apple" in Timeseries: Rethinking TSFM through the Lens of Invariance](https://mlsanity.com/paper/2510.20119/) (2025)
- [Enactive Artificial Intelligence: A Decision-Centric Architecture for Complex Systems](https://mlsanity.com/paper/2608.03413/) (2026)
- [Instance-Adaptive Prompts as Context for Time-Series Foundation Models](https://mlsanity.com/paper/2609.34786/) (2026)

## Abstract (arXiv)

Time series neural networks perform exceptionally well in real-world applications but encounter challenges such as limited scalability, poor generalization, and suboptimal zero-shot performance. Inspired by large language models, there is interest in developing large time series models (LTM) to address these issues. However, current methods struggle with training complexity, adapting human feedback, and achieving high predictive accuracy. We introduce TimeHF, a novel pipeline for creating LTMs with 6 billion parameters, incorporating human feedback. We use patch convolutional embedding to capture long time series information and design a human feedback mechanism called time-series policy optimization. Deployed in JD.com's supply chain, TimeHF handles automated replenishment for over 20,000 products, improving prediction accuracy by 33.21% over existing methods. This work advances LTM technology and shows significant industrial benefits.