# SynCast: Synergizing Contradictions in Precipitation Nowcasting via Diffusion Sequential Preference Optimization

arXiv: https://arxiv.org/abs/2510.21847 · HTML: https://mlsanity.com/paper/2510.21847/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion and multimodal model alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 31 of its 71 references that are in the map.
- First posted on arXiv 22 Oct 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Diffusion and multimodal model alignment](https://mlsanity.com/alignment/diffusion-and-multimodal-model-alignment/)

## Builds on (most cited first, 30 of 31)

- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [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)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting](https://mlsanity.com/paper/1506.04214/) (2015)
- [ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generation](https://mlsanity.com/paper/2304.05977/) (2023)
- [Training Diffusion Models with Reinforcement Learning](https://mlsanity.com/paper/2305.13301/) (2023)
- [SimPO: Simple Preference Optimization with a Reference-Free Reward](https://mlsanity.com/paper/2405.14734/) (2024)
- [Diffusion Model Alignment Using Direct Preference Optimization](https://mlsanity.com/paper/2311.12908/) (2023)
- [Directly Fine-Tuning Diffusion Models on Differentiable Rewards](https://mlsanity.com/paper/2309.17400/) (2023)
- [Skillful Precipitation Nowcasting using Deep Generative Models of Radar](https://mlsanity.com/paper/2104.00954/) (2021)
- [Aligning Text-to-Image Diffusion Models with Reward Backpropagation](https://mlsanity.com/paper/2310.03739/) (2023)
- [SimVP: Simpler yet Better Video Prediction](https://mlsanity.com/paper/2206.05099/) (2022)
- [Deep Learning for Precipitation Nowcasting: A Benchmark and A New Model](https://mlsanity.com/paper/1706.03458/) (2017)
- [PredRNN: A Recurrent Neural Network for Spatiotemporal Predictive Learning](https://mlsanity.com/paper/2103.09504/) (2021)
- [Earthformer: Exploring Space-Time Transformers for Earth System Forecasting](https://mlsanity.com/paper/2207.05833/) (2022)
- [LLM-Eval: Unified Multi-Dimensional Automatic Evaluation for Open-Domain Conversations with Large Language Models](https://mlsanity.com/paper/2305.13711/) (2023)
- [PreDiff: Precipitation Nowcasting with Latent Diffusion Models](https://mlsanity.com/paper/2307.10422/) (2023)
- [Temporal Attention Unit: Towards Efficient Spatiotemporal Predictive Learning](https://mlsanity.com/paper/2206.12126/) (2022)
- [DiffCast: A Unified Framework via Residual Diffusion for Precipitation Nowcasting](https://mlsanity.com/paper/2312.06734/) (2023)
- [CasCast: Skillful High-resolution Precipitation Nowcasting via Cascaded Modelling](https://mlsanity.com/paper/2402.04290/) (2024)
- [DisCoVQA: Temporal Distortion-Content Transformers for Video Quality Assessment](https://mlsanity.com/paper/2206.09853/) (2022)
- [Fairer Preferences Elicit Improved Human-Aligned Large Language Model Judgments](https://mlsanity.com/paper/2406.11370/) (2024)
- [Spatial Decomposition and Temporal Fusion based Inter Prediction for Learned Video Compression](https://mlsanity.com/paper/2401.15864/) (2024)
- [SPO: Multi-Dimensional Preference Sequential Alignment With Implicit Reward Modeling](https://mlsanity.com/paper/2405.12739/) (2024)
- [A Survey on Quality Metrics for Text-to-Image Generation](https://mlsanity.com/paper/2403.11821/) (2024)
- [Dissecting Human and LLM Preferences](https://mlsanity.com/paper/2402.11296/) (2024)
- [Towards Multi-dimensional Evaluation of LLM Summarization across Domains and Languages](https://mlsanity.com/paper/2506.00549/) (2025)
- [Towards Long Video Understanding via Fine-detailed Video Story Generation](https://mlsanity.com/paper/2412.06182/) (2024)

## Built on by (most cited first, 2 of 2)

- [SynWeather: Weather Observation Data Synthesis across Multiple Regions and Variables via a General Diffusion Transformer](https://mlsanity.com/paper/2511.08291/) (2025)
- [MambaRain: Multi-Scale Mamba-Attention Framework for 0-3 Hour Precipitation Nowcasting](https://mlsanity.com/paper/2605.14606/) (2026)

## Abstract (arXiv)

Precipitation nowcasting based on radar echoes plays a crucial role in monitoring extreme weather and supporting disaster prevention. Although deep learning approaches have achieved significant progress, they still face notable limitations. For example, deterministic models tend to produce over-smoothed predictions, which struggle to capture extreme events and fine-scale precipitation patterns. Probabilistic generative models, due to their inherent randomness, often show fluctuating performance across different metrics and rarely achieve consistently optimal results. Furthermore, precipitation nowcasting is typically evaluated using multiple metrics, some of which are inherently conflicting. For instance, there is often a trade-off between the Critical Success Index (CSI) and the False Alarm Ratio (FAR), making it challenging for existing models to deliver forecasts that perform well on both metrics simultaneously. To address these challenges, we introduce preference optimization into precipitation nowcasting for the first time, motivated by the success of reinforcement learning from human feedback in large language models. Specifically, we propose SynCast, a method that employs the two-stage post-training framework of Diffusion Sequential Preference Optimization (Diffusion-SPO), to progressively align conflicting metrics and consistently achieve superior performance. In the first stage, the framework focuses on reducing FAR, training the model to effectively suppress false alarms. Building on this foundation, the second stage further optimizes CSI with constraints that preserve FAR alignment, thereby achieving synergistic improvements across these conflicting metrics.