EcoAlign：一种经济理性的高效 LVLM 对齐框架
EcoAlign: An Economically Rational Framework for Efficient LVLM Alignment
把 LVLM 的推理过程建模为带预算约束的思维图搜索，用前向净现值式打分和短板原则约束路径安全，可作为视觉语言模型推理时防御越狱的可复用框架，也可用于在有限算力下权衡安全与效用的解码策略设计。
原文依据 EcoAlign incrementally expands a thought graph and scores actions using a forward-looking function
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论大型视觉语言模型（LVLM）展现出强大的推理能力，却也存在复杂的越狱漏洞。从根本上说，对齐 LVLM 不仅是安全挑战，更是一个经济效率问题。现有对齐方法难以兼顾安全性、效用与运行成本之间的权衡。尤为关键的是，仅关注最终输出（过程盲视）会把大量计算预算浪费在不安全的推理过程上。这一缺陷使得有害推理可以借助看似无害的理由加以伪装，从而绕过简单的加性安全评分。为解决该问题，我们提出 EcoAlign，一个推理时框架，它将 LVLM 视为有限理性的智能体，从而把对齐重新表述为一种经济理性的搜索。EcoAlign 增量式地扩展思维图，并使用一个前瞻性函数（类似于净现值）对动作打分，该函数会依据剩余预算动态权衡期望的安全性、效用与成本。为防止欺骗，路径安全性通过短板原则加以约束。在 6 个数据集上、跨 3 个闭源模型与 2 个开源模型的大量实验表明，EcoAlign 以更低的计算成本达到或超越最优的安全性与效用，从而为稳健的 LVLM 对齐提供了一条有原则且经济的路径。
英文原文摘要 1331 字符
Large Vision-Language Models (LVLMs) exhibit powerful reasoning capabilities but suffer sophisticated jailbreak vulnerabilities. Fundamentally, aligning LVLMs is not just a safety challenge but a problem of economic efficiency. Current alignment methods struggle with the trade-off between safety, utility, and operational costs. Critically, a focus solely on final outputs (process-blindness) wastes significant computational budget on unsafe deliberation. This flaw allows harmful reasoning to be disguised with benign justifications, thereby circumventing simple additive safety scores.To address this, we propose EcoAlign, an inference-time framework that reframes alignment as an economically rational search by treating the LVLM as a boundedly rational agent. EcoAlign incrementally expands a thought graph and scores actions using a forward-looking function (analogous to net present value) that dynamically weighs expected safety, utility, and cost against the remaining budget. To prevent deception, path safety is enforced via the weakest-link principle. Extensive experiments across 3 closed-source and 2 open-source models on 6 datasets show that EcoAlign matches or surpasses state-of-the-art safety and utility at a lower computational cost, thereby offering a principled, economical pathway to robust LVLM alignment.