Uni²: Unified Training of Latent Diffusion over Unified Modalities for De Novo Binder Design

Structure-based binder design asks a model to generate a peptide, antibody, or small molecule that binds a given target. One family of approaches handles this task with latent diffusion: an autoencoder compresses full-atom complexes into a learned latent space, and a diffusion model generates there. That latent space is usually obtained in two steps: an autoencoder is fit to reconstruct complexes, then frozen, and a diffusion model is trained on top of it. The autoencoder therefore meets the target only as data to be compressed, never as a constraint on what the generator will later have to produce, and a space that reconstructs cleanly can still hand the diffusion model a geometry that is difficult to occupy. We formulate autoencoding and latent diffusion as a unified optimization problem, jointly learning the latent representation and its generative dynamics. By allowing diffusion gradients to flow into the autoencoder, the diffusion objective directly shapes the latent manifold, aligning representation learning with the generative task while retaining the structural information required for reconstruction. On peptide, antibody, and small molecule benchmarks, our training strategy improves structural accuracy, physical plausibility, and binding energy together. These gains persist when the three modalities are trained together in one model. The change is visible in the latent space itself. Measured with a standardized sliced Wasserstein distance, the gap between posterior latents and generated latents narrows once training is unified. Finally, an ablation that treats ${\mathbf{z}}_x$ and ${\mathbf{z}}_h$ separately measures what unified training contributes through each channel: ${\mathbf{z}}_x$ accounts for the gain in geometry, ${\mathbf{z}}_h$ accounts for the gain in chemistry and interaction, and only when both receive gradients does the model become Pareto-optimal.