English(EN) Rules, Traces and Routing Statistics: Training and Calibration Data for Open Models on ARC-AGI-2

开放模型通过定制数据和推理在 ARC-AGI-2 上展现出改进的推理能力

作者 PulseAugur 编辑部 · [1 个来源] · 2026-11-05 05:00

研究人员调查了不同训练和校准数据对三个开源AI模型在ARC-AGI-2推理基准测试中的性能影响。对于NVARC Qwen3-4B模型，开发了一个包含12,000条可执行规则和95,993个任务的语料库，通过在真实任务记录上进行微调，将其得分从28.96提高到33.06。Qwen3.8-27B推理模型的得分通过增强的推理过程（包括“思考帽”和程序检查）从26.81提高到44.03。Qwen3.8-Flash-Next混合专家模型通过优化其专家层并利用带有程序检查工具的FP8索引器缓存，取得了56.67的最佳结果。 AI

影响展示了数据和推理策略如何显著提升开源AI模型的推理能力。

排序理由研究论文，详细介绍了开放模型在特定基准测试上的训练数据和推理技术。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 26 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 研究论文，详细介绍了开放模型在特定基准测试上的训练数据和推理技术。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
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: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Nguyen Minh Thang, Trinh Thi Phuong · 2026-11-05 05:00

ARC-AGI-2 上开源模型的规则、追踪和路由统计：训练与校准数据

arXiv:2611.06827v1 Announce Type: new Abstract: We study which training and calibration data help three open models solve ARC-AGI-2 tasks when the solver runs offline on four L4 GPUs for at most 12 hours. We compare data offline on public tasks and report 189 runs on the 120 hidd…

本文如何被排名

报道来源 [1]

ARC-AGI-2 上开源模型的规则、追踪和路由统计：训练与校准数据

相关实体

相关话题