# Learning an Efficient Optimizer via Hybrid-Policy Sub-Trajectory Balance

arXiv: https://arxiv.org/abs/2511.00543 · HTML: https://mlsanity.com/paper/2511.00543/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 50 references that are in the map.
- First posted on arXiv 1 Nov 2025.

## Builds on (most cited first, 27 of 27)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks](https://mlsanity.com/paper/1703.03400/) (2017)
- [Fine-Grained Visual Classification of Aircraft](https://mlsanity.com/paper/1306.5151/) (2013)
- [Sharpness-Aware Minimization for Efficiently Improving Generalization](https://mlsanity.com/paper/2010.01412/) (2020)
- [Meta-Learning for Semi-Supervised Few-Shot Classification](https://mlsanity.com/paper/1803.00676/) (2018)
- [DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition](https://mlsanity.com/paper/1310.1531/) (2013)
- [GFlowNet Foundations](https://mlsanity.com/paper/2111.09266/) (2021)
- [Trajectory balance: Improved credit assignment in GFlowNets](https://mlsanity.com/paper/2201.13259/) (2022)
- [Graph HyperNetworks for Neural Architecture Search](https://mlsanity.com/paper/1810.05749/) (2018)
- [Parameter Prediction for Unseen Deep Architectures](https://mlsanity.com/paper/2110.13100/) (2021)
- [ReAgent-V: A Reward-Driven Multi-Agent Framework for Video Understanding](https://mlsanity.com/paper/2506.01300/) (2025)
- [MetaDiff: Meta-Learning with Conditional Diffusion for Few-Shot Learning](https://mlsanity.com/paper/2307.16424/) (2023)
- [Diffusion-Based Neural Network Weights Generation](https://mlsanity.com/paper/2402.18153/) (2024)
- [Can We Scale Transformers to Predict Parameters of Diverse ImageNet Models?](https://mlsanity.com/paper/2303.04143/) (2023)
- [Enhancing Low-Cost Video Editing with Lightweight Adaptors and Temporal-Aware Inversion](https://mlsanity.com/paper/2501.04606/) (2025)
- [RAG-RLRC-LaySum at BioLaySumm: Integrating Retrieval-Augmented Generation and Readability Control for Layman Summarization of Biomedical Texts](https://mlsanity.com/paper/2405.13179/) (2024)
- [An Automated Survey of Generative Artificial Intelligence: Large Language Models, Architectures, Protocols, and Applications](https://mlsanity.com/paper/2306.02781/) (2023)
- [Deep Generative Models for Offline Policy Learning: Tutorial, Survey, and Perspectives on Future Directions](https://mlsanity.com/paper/2402.13777/) (2024)
- [Image-free Classifier Injection for Zero-Shot Classification](https://mlsanity.com/paper/2308.10599/) (2023)
- [DomainNet: Homograph Detection for Data Lake Disambiguation](https://mlsanity.com/paper/2103.09940/) (2021)
- [OCD: Learning to Overfit with Conditional Diffusion Models](https://mlsanity.com/paper/2210.00471/) (2022)
- [Adaptive Masking Enhances Visual Grounding](https://mlsanity.com/paper/2410.03161/) (2024)
- [Scene-Aware Explainable Multimodal Trajectory Prediction](https://mlsanity.com/paper/2410.16795/) (2024)

## Built on by (most cited first, 5 of 5)

- [Video Understanding: From Geometry and Semantics to Unified Models](https://mlsanity.com/paper/2603.17840/) (2026)
- [BCL: Bayesian In-Context Learning Framework for Information Extraction](https://mlsanity.com/paper/2606.18620/) (2026)
- [Grounding Multi-Hop Reasoning in Structural Causal Models via Group Relative Policy Optimization](https://mlsanity.com/paper/2605.01482/) (2026)
- [Improving Multimodal Reasoning via Worst Dimension Optimization](https://mlsanity.com/paper/2606.07801/) (2026)
- [SymDiag: Explainable Diagnosis for LLM Reasoning via Neuro-Symbolic Verification](https://mlsanity.com/paper/2608.08786/) (2026)

## Abstract (arXiv)

Recent advances in generative modeling enable neural networks to generate weights without relying on gradient-based optimization. However, current methods are limited by issues of over-coupling and long-horizon. The former tightly binds weight generation with task-specific objectives, thereby limiting the flexibility of the learned optimizer. The latter leads to inefficiency and low accuracy during inference, caused by the lack of local constraints. In this paper, we propose Lo-Hp, a decoupled two-stage weight generation framework that enhances flexibility through learning various optimization policies. It adopts a hybrid-policy sub-trajectory balance objective, which integrates on-policy and off-policy learning to capture local optimization policies. Theoretically, we demonstrate that learning solely local optimization policies can address the long-horizon issue while enhancing the generation of global optimal weights. In addition, we validate Lo-Hp's superior accuracy and inference efficiency in tasks that require frequent weight updates, such as transfer learning, few-shot learning, domain generalization, and large language model adaptation.