Training-Free Guidance Beyond Differentiability: Scalable Path Steering with Tree Search in Diffusion and Flow Models
원문 초록
Training-free guidance enables controlled generation in diffusion and flow models, but most existing methods assume differentiable objectives and rely on gradients. This work focuses on training-free guidance addressing challenges from non-differentiable objectives and discrete data distributions. We propose an algorithmic framework TreeG: Tree Search-Based Path Steering Guidance, applicable to both continuous and discrete settings in diffusion and flow models. TreeG offers a unified perspective on training-free guidance: proposing candidates for the next step, evaluating candidates, and selecting the best to move forward, enhanced by a tree search mechanism over active paths or parallelizing exploration. We comprehensively investigate the design space of TreeG over the candidate proposal module and the evaluation function, instantiating TreeG into three novel algorithms. Our experiments show that TreeG consistently outperforms the top guidance baselines in symbolic music generation, small molecule generation, and enhancer DNA design, all of which involve non-differentiable challenges. Additionally, we identify an inference-time scaling law showing TreeG's scalability in inference-time computation.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
WildFX: A DAW-Powered Pipeline for In-the-Wild Audio FX Graph Modeling
TOMI: Transforming and Organizing Music Ideas for Multi-Track Compositions with Full-Song Structure
From Generality to Mastery: Composer-Style Symbolic Music Generation via Large-Scale Pre-training
Double Entendre: Robust Audio-Based AI-Generated Lyrics Detection via Multi-View Fusion
Personalizable Long-Context Symbolic Music Infilling with MIDI-RWKV
LeVo: High-Quality Song Generation with Multi-Preference Alignment
토론
이 논문과 연결된 포럼 글이 아직 없습니다.