REVIEW 7 cited by
T2V-Turbo: Breaking the Quality Bottleneck of Video Consistency Model with Mixed Reward Feedback
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
abstract
Diffusion-based text-to-video (T2V) models have achieved significant success but continue to be hampered by the slow sampling speed of their iterative sampling processes. To address the challenge, consistency models have been proposed to facilitate fast inference, albeit at the cost of sample quality. In this work, we aim to break the quality bottleneck of a video consistency model (VCM) to achieve $\textbf{both fast and high-quality video generation}$. We introduce T2V-Turbo, which integrates feedback from a mixture of differentiable reward models into the consistency distillation (CD) process of a pre-trained T2V model. Notably, we directly optimize rewards associated with single-step generations that arise naturally from computing the CD loss, effectively bypassing the memory constraints imposed by backpropagating gradients through an iterative sampling process. Remarkably, the 4-step generations from our T2V-Turbo achieve the highest total score on VBench, even surpassing Gen-2 and Pika. We further conduct human evaluations to corroborate the results, validating that the 4-step generations from our T2V-Turbo are preferred over the 50-step DDIM samples from their teacher models, representing more than a tenfold acceleration while improving video generation quality.
Forward citations
Cited by 7 Pith papers
-
ShortFT: Diffusion Model Alignment via Shortcut-based Fine-Tuning
ShortFT fine-tunes Stable Diffusion by backpropagating reward gradients through a distilled few-step shortcut denoising chain, improving alignment scores over DRaFT-LV and DRTune.
-
When Distillation Breaks Motion Control: Restoring Generative Trajectories for Fast Video Generators
MotionEcho adaptively re-injects teacher-model guidance into few-step distilled video generators so reference motion can be copied at test time without training.
-
VQ-Insight: Teaching VLMs for AI-Generated Video Quality Understanding via Progressive Visual Reinforcement Learning
VQ-Insight uses progressive reinforcement learning with temporal shuffle and task rewards to teach a vision-language model to score and compare AI-generated videos, with gains on multiple video quality benchmarks.
-
AlignHuman: Improving Motion and Fidelity via Timestep-Segment Preference Optimization for Audio-Driven Human Animation
Timestep-segment preference optimization with separate motion and fidelity LoRAs improves audio-driven human animation quality and allows a 3.3x inference speedup.
-
From World Action Models to Embodied Brains: A Roadmap for Open-World Physical Intelligence
Physical intelligence needs an embodied brain that reasons over interventions and emits capability requests, grounded by a physical harness and shared experience contracts rather than direct actuator policies.
-
Robust Single-Stage Fully Sparse 3D Object Detection via Detachable Latent Diffusion
The submitted manuscript's abstract and full text are mismatched; the claimed 3D detection method is not present in the body.
-
Reinforcement Learning: From Algorithms To Foundation Models
A dissertation uniting the author's published results: non-exploitable Nash-DQN policies and the FightLadder benchmark for games, plus diffusion/consistency-model world models for RL — a compilation rather than new results.
Discussion (0). Sign in to comment.