Jul 2026· Visual Computing for Industry, Biomedicine, and Art· Vol 9· 0 citations· 25 references
Computer ScienceMedicine
Abstract
Text-to-motion generation aims to synthesize semantically consistent and naturally coherent motion sequences from natural language descriptions. Given the continuous nature of human motion, diffusion models operating in a continuous latent space offer inherent advantages over vector quantization-based methods, particularly in avoiding quantization errors and in modeling quality. However, existing diffusion models primarily rely on mean squared error loss. This stepwise regression paradigm often leads to ‘over-smoothed’ motion sequences and struggles to capture the subtle semantic nuances embedded in textual descriptions. To realize the potential for continuous diffusion generation, an enhanced latent-space diffusion framework designed to elevate generation capabilities across two dimensions, namely, distribution approximation and semantic alignment, is proposed. Specifically, a latent-space adversarial discriminator is incorporated. By applying decoupled adversarial supervision, this component mitigates the detail loss caused by mean regression, significantly enhancing the physical realism and dynamic sharpness. Concurrently, a latent-space contrastive alignment strategy is introduced during the denoising process that reinforces the correspondence of the generated motion sequences with the given textual inputs via explicit cross-modal constraints. Extensive experiments on standard benchmarks demonstrate that the proposed method effectively addresses the limitations of conventional diffusion models, thus validating the potential of continuous diffusion frameworks within the domain of text-driven motion synthesis.
Text-to-motion generation must produce motions that are semantically correct, temporally coherent, and physically plausible. A natural approach is to first project motion data into a structured semantic space and then train a generative model within that space. Such a paradigm has been highly successful in image generation through Representation Autoencoders (RAEs), where a frozen self-supervised encoder provides semantic features for diffusion or flow models to learn from. However, direct transfer of such a paradigm to motion space using Motion-JEPA as the frozen encoder fails dramatically. We diagnose this failure geometrically and identify two motion-specific bottlenecks: (1) the JEPA feature space is spectrally ill-conditioned, making the Gaussian-to-data transport unstable; and (2) even with a well-conditioned spectrum, flow residuals tend to align with decoder-sensitive directions, where small latent errors are amplified into large motion artifacts after decoding. Based on these insights, we propose MoRAE. MoRAE addresses the two bottlenecks separately. A compact bottleneck distills the structured JEPA representation while removing weak and redundant directions, bringing the latent spectrum into a transport-stable regime. Motion-coupled training then aligns the retained latent geometry with the decoder, making characteristic flow errors less costly after decoding. With this flow-friendly latent, a standard non-autoregressive Flow-Matching DiT achieves state-of-the-art performance.
Yifei Zhu, Mingyi Shi, Yangyang Cai et al.· 0 citations
Recent advances in diffusion models and Transformer architectures have led to significant progress in text-to-video generation. However, these models often suffer from semantic errors such as missing objects, incorrect attributes, or mismatched actions. Although some semantic correction methods perform optimization before sampling or refinement after sampling, how to detect and correct semantic deviations during the video generation process remains underexplored. In this paper, we introduce a training-free, interpretable mid-generation correction framework that integrates multimodal large language model (MLLM) feedback directly into the diffusion sampling loop. Our framework achieves diffusion trajectory correction by injecting semantic evaluation signals during video synthesis, enabling the model to optimize the generated content through continuous self-reflection. We propose two key modules: a Semantic Assessment Supervisor that generates intermediate preview frames for semantic evaluations and deviation diagnostics, and a Semantic Modification Assistant that corrects semantic drift during inference via a controllable latent trajectory intervention. Our method improves semantic alignment, visual fidelity, and temporal consistency without modifying model parameters. We validate the effectiveness of our approach through extensive experiments across multiple benchmarks.
Junhao Chen, Zheqi Lv, Keting Yin et al.· 0 citations
Motion quantization codebooks have been widely adopted to facilitate co-speech motion generation. However, the conventional quantization-based generation paradigm—which relies on probabilistic token sampling from limited discrete codebooks—suffers from two major limitations: crude, unreasonable motion representations and fixed, homogenized motion token sequences. To overcome these issues, we propose a novel explicit generation paradigm based on generative continuous quantization. Specifically, we first introduce a continuous quantization method to derive a set of generative motion units. This approach enables smoother and more accurate representation of human motion compared to classical methods. Building on these generative units, we further propose a compositional weight generation paradigm that replaces probabilistic sampling with deterministic, explicit motion synthesis. Moreover, as generalization capability is crucial for real-world deployment, we design a fully audio-aware encoder to extract style features that are decoupled from content. These features are integrated into the motion decoder via Adaptive Instance Normalization to enhance cross-speaker facial style generalization. Our method achieves state-of-the-art performance on two public datasets. Notably, owing to its concise and efficient architecture, our model attains an inference speed exceeding 4000 fps on the SHOW dataset, demonstrating strong potential for practical real-time applications.
Jialu Li, Yifan Zhao, Xin Guo et al.· IEEE Transactions on Image P...· 0 citations
Language remains an outlier in generative modeling: while images, video, and audio are increasingly modeled in continuous latent spaces, text generation still relies predominantly on discrete tokens. Existing continuous language models either inherit embedding spaces not designed for joint generation and decoding, or compress autoencoded latents to ease diffusion, sacrificing token-level fidelity. Instead of simplifying the representation to suit the generative model, we preserve a high-capacity, decodable text latent and design the diffusion model to learn its distribution directly. We introduce AURORA-LM, a continuous-latent diffusion language model that separates the construction of a decodable text representation from the modeling of its distribution. A Query-based Encoder-Decoder organizes text into a high-capacity, prefix-aligned latent sequence, and a Block-causal Diffusion Transformer learns its distribution through flow matching, generating blocks left to right while denoising positions within each block in parallel. Because such a latent is harder for diffusion to model, AURORA-LM restricts only the noisy-input pathway while retaining the full clean-latent prediction target, accommodating full-width latents without reducing decoder-facing capacity. We further calibrate the noise-level distribution to the latent width, and introduce self-trajectory consistency to bridge independently sampled training noise and iterative denoising at inference. AURORA-LM achieves the strongest performance among evaluated continuous and diffusion-based language models on OpenWebText free generation and XSum summarization. Scaling to 1B parameters with about 1500 EFLOPs of total compute yields further gains, surpassing a larger publicly released latent-diffusion language model under a matched evaluation protocol. All experiments are conducted on Ascend NPUs.
Jiajun Liang, Yu-Ling Liao, Yukang Cao et al.· 0 citations
Recent advances in video diffusion models have greatly improved visual fidelity, yet their generated motions often violate physical plausibility. We observe a common kinematic failure,"motion entanglement", the unintended coupling of independent motion sources, such as camera movement and object motion. We identify that this issue stems from data bias and the reconstruction-based training design of diffusion models. Training on noisy videos that still retain coarse motion cues inadvertently encourages the model to replicate existing motion without an incentive to learn how to model kinematically-grounded motions. To address this, we propose a Self-Imagination Fine-Tuning (SIFT) paradigm, which enables the model to learn from its own generated videos rather than directly reconstructing real ones, breaking the reconstruction shortcut. We further employ motion-aware discriminative supervision and a progressive hard-case replay strategy to stabilize and accelerate learning. By leveraging freely-generated text prompts, our method can densely cover a broad motion space, including rare or finely-disentangled scenarios that would be costly to collect as video data. Extensive experiments demonstrate that our approach substantially improves the physical realism, motion disentanglement, and controllability of generated videos.
Text-to-Audio-Video (T2AV) generation aims to produce temporally and semantically aligned visual and auditory content from natural language descriptions. While recent progress in text-to-audio and text-to-video models has improved generation quality within each modality, jointly modeling them remains challenging due to incomplete and asymmetric correspondence: audio often reflects only a subset of the visual scene, and vice versa. Naively enforcing full alignment introduces semantic noise and temporal mismatches. To address this, we propose a novel framework that performs selective cross-modal alignment through a learnable masking mechanism, enabling the model to isolate and align only the shared latent components relevant to both modalities. This mechanism is integrated into an adaptation module that interfaces with pretrained encoders and decoders from latent video and audio diffusion models, preserving their generative capacity with reduced training overhead. Theoretically, we show that our masked objective provably recovers the minimal set of shared latent variables across modalities. Empirically, our method achieves state-of-the-art performance on standard T2AV benchmarks, demonstrating significant improvements in audiovisual synchronization and semantic consistency.
Jiyang Zheng, Siqi Pan, Yu Yao et al.· Neural Information Processin...· 6 citations