CVDec 29, 2025

Task-oriented Learnable Diffusion Timesteps for Universal Few-shot Learning of Dense Tasks

arXiv:2512.23210v2h-index: 7
Originality Highly original
AI Analysis

This addresses a bottleneck in diffusion model applications for universal few-shot learning of dense tasks, offering a parameter-efficient solution.

The paper tackles the suboptimal performance in few-shot dense prediction tasks due to heuristic selection of diffusion timestep features by proposing a method to adaptively select and consolidate timesteps, achieving superior performance on the Taskonomy dataset with only a few support queries.

Denoising diffusion probabilistic models have brought tremendous advances in generative tasks, achieving state-of-the-art performance thus far. Current diffusion model-based applications exploit the power of learned visual representations from multistep forward-backward Markovian processes for single-task prediction tasks by attaching a task-specific decoder. However, the heuristic selection of diffusion timestep features still heavily relies on empirical intuition, often leading to sub-optimal performance biased towards certain tasks. To alleviate this constraint, we investigate the significance of versatile diffusion timestep features by adaptively selecting timesteps best suited for the few-shot dense prediction task, evaluated on an arbitrary unseen task. To this end, we propose two modules: Task-aware Timestep Selection (TTS) to select ideal diffusion timesteps based on timestep-wise losses and similarity scores, and Timestep Feature Consolidation (TFC) to consolidate the selected timestep features to improve the dense predictive performance in a few-shot setting. Accompanied by our parameter-efficient fine-tuning adapter, our framework effectively achieves superiority in dense prediction performance given only a few support queries. We empirically validate our learnable timestep consolidation method on the large-scale challenging Taskonomy dataset for dense prediction, particularly for practical universal and few-shot learning scenarios.

Foundations

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

Your Notes