PulseAugur
中
实时 21:22:20
English(EN) Agon: Competitive Cross-Model RL with Implicit Rival Grading of Reasoning

Agon 框架使用竞争性 AI 模型对推理进行评分

研究人员推出了一种新颖的竞争性强化学习框架 Agon,旨在提高 AI 模型的推理能力。与仅对最终答案评分的传统方法不同,Agon 将两个模型进行对抗,每个模型通过隐式方式对另一个模型的推理过程进行评分。这种竞争性设置通过面对日益强大的对手,迫使模型发展出更好的思考策略,从而带来显著的性能提升。在 DeepMath 数据集上使用 Qwen3 进行测试时,Agon 的 pass@1 率是标准 GRPO 的两倍,并且比未经训练的 Mixture-of-Agents 方法提高了八倍。 AI

影响 这种竞争性训练方法可能带来更强大、更具能力的推理模型,从而加速在复杂问题解决任务中的进展。

排序理由 该集群包含一篇详细介绍 AI 模型训练新研究方法的学术论文。

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

Agon 框架使用竞争性 AI 模型对推理进行评分

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
该集群包含一篇详细介绍 AI 模型训练新研究方法的学术论文。
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
paper, model release
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
92 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [2]

  1. arXiv cs.AI TIER_1 English(EN) · Vladislav Beliaev ·

    Agon:具有隐式竞争对手评分的竞争性跨模型强化学习

    arXiv:2607.07690v1 Announce Type: cross Abstract: Reinforcement learning from verifiable rewards (e.g. GRPO) is the engine behind today's reasoning models, yet it grades only the final answer. On hard problems this trains models to write more rather than to think better, since th…

  2. arXiv cs.AI TIER_1 English(EN) · Vladislav Beliaev ·

    Agon:具有隐式对手评分的竞争性跨模型强化学习

    Reinforcement learning from verifiable rewards (e.g. GRPO) is the engine behind today's reasoning models, yet it grades only the final answer. On hard problems this trains models to write more rather than to think better, since the trace itself is never graded and no label for go…