English(EN) Algorithmic Scratchpads and Curriculum Staging for Arithmetic Reasoning in Tiny Transformers

小型Transformer通过新的训练方法实现更高的算术准确性

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-08 04:00

研究人员开发了一种方法，通过使用算法草稿板和分层课程来提高小型语言模型的算术推理能力。他们发现，正确的数据加载和语言预训练至关重要，而像RoPE和SwiGLU这样的现代架构组件可以提高性能。特定的逐位长除法草稿板显著提高了准确性，但由于加法复杂性，多位数乘法仍然具有挑战性。该研究还强调了在无缓冲训练期间未见操作数和灾难性遗忘的问题。 AI

影响展示了提高小型、更高效语言模型推理能力的技术。

排序理由学术论文，详细介绍了小型语言模型在算术任务上的新颖训练方法。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 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: 7 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 学术论文，详细介绍了小型语言模型在算术任务上的新颖训练方法。[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: Same-day; Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.LG TIER_1 English(EN) · Sourabh Kasliwal · 2026-10-08 04:00

Tiny Transformers 中的算法草稿板和课程分阶段用于算术推理

arXiv:2610.09003v1 Announce Type: new Abstract: Autoregressive Large Language Models (LLMs) frequently struggle with deterministic multi-step algorithmic tasks such as multi-digit multiplication and long division. In this paper, we investigate the mechanics of multi-step arithmet…

本文如何被排名

报道来源 [1]

Tiny Transformers 中的算法草稿板和课程分阶段用于算术推理

相关实体

相关话题