---
title: "Multi-Objective Generative Optimization via Expected Hypervolume Fine-Tuning"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.05207v1"
permalink: "https://feed.astralor.com/items/itm_20c575b84680ce89392f511f"
published: "2026-11-03T16:55:58.000Z"
score: 61
channel: "tech"
tags: ["推理效率","RAG","Multi-Objective Generative Optimization","Expected Hypervolume Fine-Tuning","Scientific","Pretrained","Inference-time","Hypervolume Fine-Tuning"]
---

# Multi-Objective Generative Optimization via Expected Hypervolume Fine-Tuning

## At a glance

Scientific discovery requires plausible designs spanning useful trade-offs among competing properties: a molecule with strong binding affinity, for example, may be difficult to synthesize. Pretrained generative models, …

## Key points

- Scientific discovery requires plausible designs spanning useful trade-offs among competing properties: a molecule with strong binding affinity, for example, may be difficult to synthesize.
- Pretrained generative models, e.g., flow and diffusion generators, provide a foundation for proposing such designs, but steering them toward high-quality, complementary trade-offs remains challenging.
- Inference-time search incurs high per-sample costs, while fine-tuning typically relies on fixed per-candidate rewards that do not capture how designs complement one another.

## Full text

Scientific discovery requires plausible designs spanning useful trade-offs among competing properties: a molecule with strong binding affinity, for example, may be difficult to synthesize. Pretrained generative models, e.g., flow and diffusion generators, provide a foundation for proposing such designs, but steering them toward high-quality, complementary trade-offs remains challenging. Inference-time search incurs high per-sample costs, while fine-tuning typically relies on fixed per-candidate rewards that do not capture how designs complement one another. We introduce Hypervolume Fine-Tuning (HVFT), which adapts a single pretrained generator to maximize the expected hypervolume of its samples while controlling departure from the pretrained distribution. Expected hypervolume jointly rewards candidate quality and collective trade-off coverage. Nonetheless, we show that it cannot be expressed as the expectation of a fixed per-candidate reward and therefore cannot be optimized directly with standard techniques. A key observation of this work is that its first variation corresponds to the expected marginal hypervolume contribution of a candidate, which yields a tractable adaptive reward via an unbiased estimator based on model samples. HVFT uses this reward to decompose the objective into a sequence of KL-proximal scalar-reward subproblems. We establish concavity, relative smoothness, and convergence guarantees for the resulting distributional procedure. Across therapeutic-peptide and molecular design tasks spanning discrete diffusion and continuous flow-matching models, HVFT achieves state-of-the-art expected hypervolume and trade-off coverage.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_20c575b84680ce89392f511f