Wan2.1-Fun-Reward-LoRAs

Introduction

We explore the Reward Backpropagation technique ^{1 2} to optimized the generated videos by Wan2.1-Fun for better alignment with human preferences. We provide the following pre-trained models (i.e. LoRAs) along with the training script. You can use these LoRAs to enhance the corresponding base model as a plug-in or train your own reward LoRA.

For more details, please refer to our GitHub repo.

Name	Base Model	Reward Model	Hugging Face	Description
Wan2.1-Fun-1.3B-InP-HPS2.1.safetensors	Wan2.1-Fun-1.3B-InP	HPS v2.1	🤗Link	Official HPS v2.1 reward LoRA (`rank=128` and `network_alpha=64`) for Wan2.1-Fun-1.3B-InP. It is trained with a batch size of 8 for 5,000 steps.
Wan2.1-Fun-1.3B-InP-MPS.safetensors	Wan2.1-Fun-1.3B-InP	MPS	🤗Link	Official MPS reward LoRA (`rank=128` and `network_alpha=64`) for Wan2.1-Fun-1.3B-InP. It is trained with a batch size of 8 for 7,500 steps.
Wan2.1-Fun-14B-InP-HPS2.1.safetensors	Wan2.1-Fun-14B-InP	HPS v2.1	🤗Link	Official HPS v2.1 reward LoRA (`rank=128` and `network_alpha=64`) for Wan2.1-Fun-14B-InP. It is trained with a batch size of 32 for 3,000 steps.
Wan2.1-Fun-14B-InP-MPS.safetensors	Wan2.1-Fun-14B-InP	MPS	🤗Link	Official MPS reward LoRA (`rank=128` and `network_alpha=64`) for Wan2.1-Fun-14B-InP. It is trained with a batch size of 8 for 4,500 steps.

Demo

Wan2.1-Fun-1.3B-InP

Prompt	Wan2.1-Fun-1.3B-InP	Wan2.1-Fun-1.3B-InP HPSv2.1 Reward LoRA	Wan2.1-Fun-1.3B-InP MPS Reward LoRA
A kangaroo bounds across the plain and a cow grazes Expanded In a vast, sun-drenched Australian plain, a lively kangaroo bounds with powerful leaps across the dry grass, its shadow following closely. Nearby, a serene brown and white cow grazes leisurely, its tail swishing gently in the warm breeze. The sky is a vibrant blue, dotted with fluffy clouds.
A penguin waddles on the ice, a camel treks by Expanded A small penguin waddles slowly across a vast, icy surface under a clear blue sky. The penguin's short, flipper-like wings sway at its sides as it moves. Nearby, a camel treks steadily, its long legs navigating the snowy terrain with ease. The camel's fur is thick, providing warmth in the cold environment.
Porcelain rabbit hopping by a golden cactus Expanded A delicate porcelain rabbit, with intricate painted details, hops gracefully across a sandy desert floor. Nearby, a golden cactus stands tall, its metallic surface glimmering in the sunlight. The backdrop features rolling sand dunes under a clear blue sky, casting gentle shadows.
Pig with wings flying above a diamond mountain Expanded A whimsical pig, complete with delicate feathered wings, soars gracefully above a shimmering diamond mountain. The pig's pink skin glistens in the sunlight as it flaps its wings. The mountain below sparkles with countless facets, reflecting brilliant rays of light into the clear blue sky.

[!NOTE] The above test prompts are from T2V-CompBench and expanded into detailed prompts by Llama-3.3. Videos are generated with HPSv2.1 Reward LoRA weight 0.5 and MPS Reward LoRA weight 0.7.

Wan2.1-Fun-14B-InP

Prompt	Wan2.1-Fun-1.3B-InP	Wan2.1-Fun-1.3B-InP HPSv2.1 Reward LoRA	Wan2.1-Fun-1.3B-InP MPS Reward LoRA
A panda eats bamboo while a monkey swings from branch to branch Expanded In a lush green forest, a panda sits comfortably against a tree, leisurely munching on bamboo stalks. Nearby, a lively monkey swings energetically from branch to branch, its tail curling around the limbs. Sunlight filters through the canopy, casting dappled shadows on the forest floor.
A dog runs through a field while a cat climbs a tree Expanded In a sunlit, expansive green field surrounded by tall trees, a playful golden retriever sprints energetically across the grass, its fur gleaming in the afternoon sun. Nearby, a nimble tabby cat gracefully climbs a sturdy tree, its claws gripping the bark effortlessly. The sky is clear blue with occasional birds flying.
Elderly artist with a white beard painting on a white canvas Expanded An elderly artist with a long white beard stands in a sunlit studio surrounded by art supplies. He wears a paint-splattered apron over a casual shirt. His hand moves gracefully as he paints vibrant colors on a large white canvas positioned on an easel. The studio is filled with natural light streaming through tall windows, highlighting the textures of his work.
Pig with wings flying above a diamond mountain Expanded A whimsical pig, complete with delicate feathered wings, soars gracefully above a shimmering diamond mountain. The pig's pink skin glistens in the sunlight as it flaps its wings. The mountain below sparkles with countless facets, reflecting brilliant rays of light into the clear blue sky.

[!NOTE] The above test prompts are from T2V-CompBench and expanded into detailed prompts by Llama-3.3. Videos are generated with HPSv2.1 Reward LoRA weight 0.7 and MPS Reward LoRA weight 0.7.

Quick Start

Set lora_path and lora_weight in examples/wan2.1_fun/predict_t2v.py.

Limitations

We observe after training to a certain extent, the reward continues to increase, but the quality of the generated videos does not further improve. The model trickly learns some shortcuts (by adding artifacts in the background, i.e., adversarial patches) to increase the reward.
Currently, there is still a lack of suitable preference models for video generation. Directly using image preference models cannot evaluate preferences along the temporal dimension (such as dynamism and consistency). Further more, We find using image preference models leads to a decrease in the dynamism of generated videos. Although this can be mitigated by computing the reward using only the first frame of the decoded video, the impact still persists.

Reference

Clark, Kevin, et al. "Directly fine-tuning diffusion models on differentiable rewards.". In ICLR 2024.
Prabhudesai, Mihir, et al. "Aligning text-to-image diffusion models with reward backpropagation." arXiv preprint arXiv:2310.03739 (2023).

alibaba-pai/Wan2.1-Fun-Reward-LoRAs