揭秘LLM中隐藏的“超级权重”：为何选择性训练策略翻车？COLM 2026接受研究揭示深层原理。
arXiv:2607.08733v1 Announce Type: new Abstract: Recent work identified Super Weights, individual parameters whose removal degrades model performance b…
挑战LLM强化学习传统认知，揭示单调推理策略才是真正优化目标而非训练策略。
arXiv:2606.29526v1 Announce Type: new Abstract: Reinforcement learning (RL) has gained growing attention in large language model (LLM) post-training, …
一篇剖析强化学习更新大模型推理能力的关键因素研究，揭示影响性能的核心变量与训练策略。
arXiv:2606.22570v1 Announce Type: new Abstract: Reinforcement Learning from Verifiable Rewards (RLVR) has emerged as a promising framework for enhanci…
探究LLM在生成式推荐中的记忆机制，揭示其对推荐系统泛化能力的影响与优化训练策略
arXiv:2606.17276v1 Announce Type: cross Abstract: Generative recommendation (GR) has emerged as a promising direction for recommender systems. Recentl…
Piper 让你用少量注解自定义分布式训练策略，真正实现策略与运行时的解耦。
arXiv:2606.11169v1 Announce Type: cross Abstract: Large-scale model training increasingly relies on composing multiple parallelism strategies, such as…
白质高信号与卒中病变分割中，如何用部分标注数据优化训练策略？这篇论文系统性对比了多种方法。
arXiv:2601.20503v2 Announce Type: replace Abstract: White matter hyperintensities (WMH) and ischaemic stroke lesions (ISL) are key imaging biomarkers …
探讨SFT如何优化自身并为强化学习做准备，揭示大模型训练策略的关键演进方向
arXiv:2602.01058v2 Announce Type: replace-cross Abstract: Post-training of reasoning LLMs is a holistic process that typically consists of an offline …
ICML 2025收录，揭示数据质量如何决定大模型损失与缩放定律的深层关系。
arXiv:2502.12120v3 Announce Type: replace Abstract: Scaling laws guide the development of large language models (LLMs) by offering estimates for the o…
Google内部定制LLM实战：万亿token数据集+中训策略，专攻企业软件工程场景。
Article URL: https://arxiv.org/abs/2605.16517 Comments URL: https://news.ycombinator.com/item?id=48202484 Points: 1 # Comments: 0