# FlowPortrait: Reinforcement Learning for Audio-Driven Portrait Video Generation

arXiv: https://arxiv.org/abs/2603.00159 · HTML: https://mlsanity.com/paper/2603.00159/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in RL training of language and agent models, one of 15 directions in Deep reinforcement learning methods.
- Builds on 24 of its 32 references that are in the map.
- First posted on arXiv 25 Feb 2026.

## Placed in

- Deep reinforcement learning methods / [RL training of language and agent models](https://mlsanity.com/deeprl/rl-training-of-language-and-agent-models/)

## Builds on (most cited first, 24 of 24)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [CLIPScore: A Reference-free Evaluation Metric for Image Captioning](https://mlsanity.com/paper/2104.08718/) (2021)
- [RAFT: Recurrent All-Pairs Field Transforms for Optical Flow](https://mlsanity.com/paper/2003.12039/) (2020)
- [Towards Accurate Generative Models of Video: A New Metric & Challenges](https://mlsanity.com/paper/1812.01717/) (2018)
- [Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesis](https://mlsanity.com/paper/2306.09341/) (2023)
- [A Lip Sync Expert Is All You Need for Speech to Lip Generation In The Wild](https://mlsanity.com/paper/2008.10010/) (2020)
- [The Perception-Distortion Tradeoff](https://mlsanity.com/paper/1711.06077/) (2017)
- [Judging the Judges: A Systematic Study of Position Bias in LLM-as-a-Judge](https://mlsanity.com/paper/2406.07791/) (2024)
- [EchoMimic: Lifelike Audio-Driven Portrait Animations through Editable Landmark Conditions](https://mlsanity.com/paper/2407.08136/) (2024)
- [MixGRPO: Unlocking Flow-based GRPO Efficiency with Mixed ODE-SDE](https://mlsanity.com/paper/2507.21802/) (2025)
- [Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models](https://mlsanity.com/paper/2411.04996/) (2024)
- [Interacting particle solutions of Fokker-Planck equations through gradient-log-density estimation](https://mlsanity.com/paper/2006.00702/) (2020)
- [MetaPortrait: Identity-Preserving Talking Head Generation with Fast Personalized Adaptation](https://mlsanity.com/paper/2212.08062/) (2022)
- [Coefficients-Preserving Sampling for Reinforcement Learning with Flow Matching](https://mlsanity.com/paper/2509.05952/) (2025)
- [GRPO-Guard: Mitigating Implicit Over-Optimization in Flow Matching via Regulated Clipping](https://mlsanity.com/paper/2510.22319/) (2025)
- [High-Fidelity and Freely Controllable Talking Head Video Generation](https://mlsanity.com/paper/2304.10168/) (2023)
- [A Light Weight Model for Active Speaker Detection](https://mlsanity.com/paper/2303.04439/) (2023)
- [MEMO: Memory-Guided Diffusion for Expressive Talking Video Generation](https://mlsanity.com/paper/2412.04448/) (2024)
- [Kling-Avatar: Grounding Multimodal Instructions for Cascaded Long-Duration Avatar Animation Synthesis](https://mlsanity.com/paper/2509.09595/) (2025)

## Built on by (most cited first, 1 of 1)

- [GDPO-Listener: Expressive Interactive Head Generation via Auto-Regressive Flow Matching and Group reward-Decoupled Policy Optimization](https://mlsanity.com/paper/2603.25020/) (2026)

## Abstract (arXiv)

Generating realistic talking-head videos remains challenging due to persistent issues such as imperfect lip synchronization, unnatural motion, and evaluation metrics that correlate poorly with human perception. We propose FlowPortrait, a reinforcement-learning framework for audio-driven portrait animation built on a multimodal backbone for autoregressive audio-to-video generation. FlowPortrait introduces a human-aligned evaluation system based on Multimodal Large Language Models (MLLMs) to assess lip-sync accuracy, expressiveness, and motion quality. These signals are combined with perceptual and temporal consistency regularizers to form a stable composite reward, which is used to post-train the generator via Group Relative Policy Optimization (GRPO). Extensive experiments, including both automatic evaluations and human preference studies, demonstrate that FlowPortrait consistently produces higher-quality talking-head videos, highlighting the effectiveness of reinforcement learning for portrait animation.