Test-Time Instance-Specific Parameter Composition: A New Paradigm for Adaptive Generative Modeling

2026-03-29

AI Summary

This work addresses the limitation of existing generative models, which rely on fixed pre-trained parameters and lack the ability to dynamically adapt to individual input instances. To overcome this, the authors propose Composer, a framework that enables instance-level adaptation at test time by conditionally generating and injecting lightweight parameters based on the input, without requiring fine-tuning. Composer introduces, for the first time, a test-time mechanism for instance-specific parameter composition, endowing static models with context-awareness and dynamic adaptability. The approach is compatible with both diffusion and autoregressive architectures and supports quantized deployment. Experimental results demonstrate that Composer consistently enhances generation quality across diverse tasks while maintaining low computational and memory overhead, confirming its effectiveness and broad applicability.

Abstract

Existing generative models, such as diffusion and auto-regressive networks, are inherently static, relying on a fixed set of pretrained parameters to handle all inputs. In contrast, humans flexibly adapt their internal generative representations to each perceptual or imaginative context. Inspired by this capability, we introduce Composer, a new paradigm for adaptive generative modeling based on test-time instance-specific parameter composition. Composer generates input-conditioned parameter adaptations at inference time, which are injected into the pretrained model's weights, enabling per-input specialization without fine-tuning or retraining. Adaptation occurs once prior to multi-step generation, yielding higher-quality, context-aware outputs with minimal computational and memory overhead. Experiments show that Composer substantially improves performance across diverse generative models and use cases, including lightweight/quantized models and test-time scaling. By leveraging input-aware parameter composition, Composer establishes a new paradigm for designing generative models that dynamically adapt to each input, moving beyond static parameterization.

Problem

• adaptive generative modeling
• test-time adaptation
• instance-specific
• parameter composition
• static generative models

Innovation

• test-time adaptation
• instance-specific parameter composition
• adaptive generative modeling
• parameter injection
• dynamic model specialization