测试时实例特定参数组合：自适应生成建模的新范式
Test-Time Instance-Specific Parameter Composition: A New Paradigm for Adaptive Generative Modeling
在推理前一次性生成输入条件化的权重增量并注入预训练生成模型，为视觉生成模型提供免微调的逐样本自适应手段，可直接用于轻量化/量化扩散模型的质量补偿与测试时扩展。
原文依据 Composer generates input-conditioned parameter adaptations at inference time, which are injected into the pretrained model's weights
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论现有的生成模型（如扩散模型和自回归网络）本质上是静态的，依赖一组固定的预训练参数来处理所有输入。相比之下，人类能够灵活地调整其内部生成表征以适应每一个感知或想象的情境。受这一能力启发，我们提出 Composer，一种基于测试时实例特定参数组合的自适应生成建模新范式。Composer 在推理时生成以输入为条件的参数自适应，并将其注入预训练模型的权重中，从而在无需微调或重新训练的情况下实现针对每个输入的专门化。自适应在多步生成之前仅发生一次，以极小的计算与内存开销产生更高质量、更具上下文感知的输出。实验表明，Composer 在多种生成模型和使用场景中均显著提升性能，包括轻量化/量化模型以及测试时扩展。通过利用输入感知的参数组合，Composer 为设计能够动态适应每个输入的生成模型建立了新范式，超越了静态参数化。代码将发布于 https://github.com/tmtuan1307/Composer。
英文原文摘要 1241 字符
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. The code will be available at https://github.com/tmtuan1307/Composer.