生成对抗循环（Generative Adversarial Loops, GAL）

中文摘要 AI 辅助

AI研究进展可视为基准创建与方法发现两个过程的交互，历史上两者均由人类智能驱动。然而，近期AI进展已推动自动化方法发现加速，而自动化基准创建却受关注较少。为实现自推进系统，我们提出生成对抗循环（Generative Adversarial Loop, GAL），这是一种生成器-判别器框架，在两个智能体搜索间交替运行：（1）判别器生成对抗性数据以暴露当前系统的弱点；（2）生成器发现算法以克服这些弱点。我们将该框架应用于高效推理的近似算法。与现有主要聚焦算法发现的自动研究系统不同，GAL引入了一个判别器智能体，通过持续搜索当前算法的弱点来自动设定目标。我们在四个任务上展示了对抗性数据生成：KV压缩、稀疏视频生成、稀疏注意力和上下文扩展，其中判别器识别出了当前最先进技术的弱点。我们进一步证明GAL可实现自主改进，新发现的算法不仅在对抗性生成数据上表现更好，还在既定基准上有所提升。具体而言，GAL在KV压缩任务上以Qwen3-4B将CompactorPress提升4倍，使判别器数据集上的性能从0.35升至0.97，同时优于RULER-HARD（+0.77点）；在上下文扩展任务上，GAL使判别器数据集上的Dual Chunk Attention从0.20提升至0.90，同时在标准基准（ScienceFiction提升6点、PG19 32K困惑度降低0.33）上也取得增益。因此，GAL为自主设定目标和算法改进提供了一条路径，AI系统可在此过程中持续发现自身弱点并开发克服方法。

英文摘要

AI research progress can be viewed as the interaction between two processes: benchmark creation and method discovery. Historically, both were driven by human intelligence. However, recent advances in AI have accelerated automated method discovery, while automated benchmark creation has received comparatively less attention. To enable self-advancing systems, we propose Generative Adversarial Loop (GAL), a generator-discriminator framework alternating between two agentic searches: (1) a discriminator that generates adversarial data to expose weaknesses in current systems, and (2) a generator that discovers algorithms to overcome them. We apply this framework to approximation algorithms for efficient inference. Unlike existing auto research systems, which primarily focus on algorithm discovery, GAL introduces a discriminator agent that automates goalpost setting by continually searching for weaknesses in the current algorithm. We demonstrate adversarial data generation across four tasks: KV compression, sparse video generation, sparse attention, and context extension, where the discriminator identifies weaknesses in state of the art techniques. We further show that GAL enables autonomous improvement, with newly discovered algorithms improving not only on adversarially generated data, but also on established benchmarks. Specifically, GAL improves CompactorPress on KV compression with Qwen3-4B at 4x, raising performance on the discriminator dataset from 0.35 to 0.97, while also outperforming RULER-HARD (+0.77 pts). For context extension, GAL boosts Dual Chunk Attention from 0.20 to 0.90 on the discriminator dataset, while yielding gains on standard benchmarks(ScienceFiction (+6 pts) and PG19 32K (-0.33 PPL)). GAL thus provides a path toward autonomous goalpost setting and algorithmic improvement, where AI systems continually discover their own weaknesses and develop methods to overcome them.

发表机构

• Indian Institute of Technology Bombay（印度理工学院孟买分校）
• University of California, Berkeley（加州大学伯克利分校）

机构由 AI 辅助整理，请以论文原文为准。