AIMay 16

TILT: Improving Compositional Generation in Diffusion Models with a Model-Intrinsic Reward

arXiv:2607.21606
Originality Incremental advance
AI Analysis

This work addresses the problem of compositional failures in text-to-image generation for users requiring precise multi-concept prompts, offering a training-free method that improves alignment while maintaining image quality.

TILT improves compositional generation in diffusion models by using a model-intrinsic reward to align sampling trajectories with complex prompts, achieving better compositional alignment on T2ICompBench without external supervision.

Recent advances in powerful text-to-image generation models have made it increasingly important to develop test-time methods that modify the sampling trajectory to produce images more faithful to complex compositional prompts. We present TILT, a training-free framework for compositional text-to-image generation via test-time reward alignment. We interpret compositional failures as overlap modes between joint and single-concept distributions, and define a reward that favors samples where all concepts are jointly present. This reward is intrinsic to the base model and does not require any external supervision or reward models. This yields a KL-constrained objective with a closed-form tilted target distribution and principled guiding steps for diffusion sampling. The interaction of concept distributions together with the above reward naturally leads to two different guidance strategies while a hybrid approach that balances their respective benefits produces stronger performance. Experiments on prompts from T2ICompBench show that our method improves compositional alignment while preserving image quality compared to previous baselines.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes