arXiv：cs.AI· Kislay Aditya Oj, Nidhi Jain, Sri Surya Varma Datla, Priyanka Jayaswal, Kumar Krishna Agrawal, Aditya Desai·· 3 小时前
Generative Adversarial Loop（GAL）：让 AI 自动发现自身弱点并改进算法
Generative Adversarial Loops
AI 导读
研究者提出 Generative Adversarial Loop（GAL），一个生成器-判别器框架，通过交替进行智能体搜索来自动化基准设定与算法发现。判别器智能体生成对抗数据以暴露当前系统弱点，生成器则发现算法予以克服。
正文
Abstract: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.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.11458 [cs.LG] |
| (or arXiv:2610.11458v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.11458 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Nidhi Jain [view email]
[v1]
Thu, 8 Oct 2026 08:11:37 UTC (3,558 KB)
来源：arXiv：cs.AI · arxiv.org