# Incentivizing Generative Zero-Shot Learning via Outcome-Reward Reinforcement Learning with Visual Cues

arXiv: https://arxiv.org/abs/2603.21138 · HTML: https://mlsanity.com/paper/2603.21138/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL with diffusion and flow models, one of 15 directions in Deep reinforcement learning methods.
- Builds on 56 of its 85 references that are in the map.
- First posted on arXiv 22 Mar 2026.

## Placed in

- Deep reinforcement learning methods / [RL with diffusion and flow models](https://mlsanity.com/deeprl/rl-with-diffusion-and-flow-models/)

## Builds on (most cited first, 30 of 56)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [Tackling the Generative Learning Trilemma with Denoising Diffusion GANs](https://mlsanity.com/paper/2112.07804/) (2021)
- [Visual-RFT: Visual Reinforcement Fine-Tuning](https://mlsanity.com/paper/2503.01785/) (2025)
- [Is DPO Superior to PPO for LLM Alignment? A Comprehensive Study](https://mlsanity.com/paper/2404.10719/) (2024)
- [Feature Generating Networks for Zero-Shot Learning](https://mlsanity.com/paper/1712.00981/) (2017)
- [f-VAEGAN-D2: A Feature Generating Framework for Any-Shot Learning](https://mlsanity.com/paper/1903.10132/) (2019)
- [Text-to-Image Diffusion Models are Zero-Shot Classifiers](https://mlsanity.com/paper/2303.15233/) (2023)
- [LLM Post-Training: A Deep Dive into Reasoning Large Language Models](https://mlsanity.com/paper/2502.21321/) (2025)
- [Contrastive Embedding for Generalized Zero-Shot Learning](https://mlsanity.com/paper/2103.16173/) (2021)
- [Multi-modal Cycle-consistent Generalized Zero-Shot Learning](https://mlsanity.com/paper/1808.00136/) (2018)
- [Visual Planning: Let's Think Only with Images](https://mlsanity.com/paper/2505.11409/) (2025)
- [Leveraging the Invariant Side of Generative Zero-Shot Learning](https://mlsanity.com/paper/1904.04092/) (2019)
- [FREE: Feature Refinement for Generalized Zero-Shot Learning](https://mlsanity.com/paper/2107.13807/) (2021)
- [Leveraging Seen and Unseen Semantic Relationships for Generative Zero-Shot Learning](https://mlsanity.com/paper/2007.09549/) (2020)
- [HSVA: Hierarchical Semantic-Visual Adaptation for Zero-Shot Learning](https://mlsanity.com/paper/2109.15163/) (2021)
- [VisualPuzzles: Decoupling Multimodal Reasoning Evaluation from Domain Knowledge](https://mlsanity.com/paper/2504.10342/) (2025)
- [TransZero: Attribute-guided Transformer for Zero-Shot Learning](https://mlsanity.com/paper/2112.01683/) (2021)
- [I2MVFormer: Large Language Model Generated Multi-View Document Supervision for Zero-Shot Image Classification](https://mlsanity.com/paper/2212.02291/) (2022)
- [Improving Zero-Shot Generalization for CLIP with Synthesized Prompts](https://mlsanity.com/paper/2307.07397/) (2023)
- [MSDN: Mutually Semantic Distillation Network for Zero-Shot Learning](https://mlsanity.com/paper/2203.03137/) (2022)
- [DUET: Cross-modal Semantic Grounding for Contrastive Zero-shot Learning](https://mlsanity.com/paper/2207.01328/) (2022)
- [TransZero++: Cross Attribute-Guided Transformer for Zero-Shot Learning](https://mlsanity.com/paper/2112.08643/) (2021)
- [VGSE: Visually-Grounded Semantic Embeddings for Zero-Shot Learning](https://mlsanity.com/paper/2203.10444/) (2022)
- [I2DFormer: Learning Image to Document Attention for Zero-Shot Image Classification](https://mlsanity.com/paper/2209.10304/) (2022)

## Abstract (arXiv)

Recent advances in zero-shot learning (ZSL) have demonstrated the potential of generative models. Typically, generative ZSL synthesizes visual features conditioned on semantic prototypes to model the data distribution of unseen classes, followed by training a classifier on the synthesized data. However, the synthesized features often remain task-agnostic, leading to degraded performance. Moreover, inferring a faithful distribution from semantic prototypes alone is insufficient for classes that are semantically similar but visually distinct. To address these and advance ZSL, we propose RLVC, an outcome-reward reinforcement learning RL framework with visual cues for generative ZSL. At its core, RL empowers the generative model to self-evolve, implicitly enhancing its generation capability. In particular, RLVC updates the generative model using an outcome-based reward, encouraging the synthesis of task-relevant features. Furthermore, we introduce class-wise visual cues that (i) align synthesized features with visual prototypes and (ii) stabilize the RL training updates. For the training process, we present a novel cold-start strategy. Comprehensive experiments and analyses on three prevalent ZSL benchmarks demonstrate that RLVC achieves state-of-the-art results with a 4.7% gain.