Search papers, labs, and topics across Lattice.
AURORA-LM introduces a novel continuous-latent diffusion model for language generation that maintains high-capacity, decodable text representations while directly learning their distribution. By employing a Query-based Encoder-Decoder and a Block-causal Diffusion Transformer, the model effectively organizes text into a prefix-aligned latent sequence and generates outputs through parallel denoising. This approach not only preserves token-level fidelity but also outperforms existing models on benchmarks like OpenWebText and XSum, achieving superior performance even at a smaller scale compared to larger latent-diffusion models.
AURORA-LM achieves state-of-the-art performance in language generation by leveraging a continuous-latent diffusion framework that preserves text fidelity while enhancing generative capabilities.
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.
