热点
"LLM Training" 相关文章
Open-weight training practices and implications for CoT monitorability
少点错误 2025-11-04T11:20:45.000000Z
让AI学习效率飙升50倍的秘密：在线策略蒸馏
硅星GenAI 2025-10-28T13:33:42.000000Z
Chinese AI Startup DeepSeek Unveils Open-Source Optical Compression Model for LLM Training
钛媒体：引领未来商业与生活新知 2025-10-22T08:13:49.000000Z
不改超参、不调token：用分位数替代均值，QAE让大模型强化学习更稳定
PaperWeekly 2025-10-21T14:54:07.000000Z
豆包是如何炼成的？字节放出自研万卡训练系统ByteRobust论文
机器之心 2025-10-21T14:51:17.000000Z
不改超参、不调token：用分位数替代均值，QAE让大模型强化学习更稳定
PaperWeekly 2025-10-21T14:00:29.000000Z
[分享发现] [兑换码分发] 开箱即用、全链路功能覆盖的低代码在线大模型训练与微调服务平台｜共 20 个
V2EX 2025-10-11T08:03:20.000000Z
任意Agent皆可强化学习！微软推出Agent Lightning框架，无需修改任何代码
新智元 2025-10-10T10:05:20.000000Z
Inoculation prompting: Instructing models to misbehave at train-time can improve run-time behavior
少点错误 2025-10-08T22:19:53.000000Z
颠覆大模型后训练！陈丹琦团队提出「基于模型奖励思维的强化学习」RLMT
智源社区 2025-09-30T13:33:18.000000Z
$1.5 Billion Settlement Shakes Up AI: Anthropic Strikes Deal with Authors in Copyright Clash
AI 2 People 2025-09-29T13:53:55.000000Z
普林斯顿陈丹琦组新作：RLHF难支撑，RLVR有边界？RLMT开辟第三条路
PaperWeekly 2025-09-27T01:08:13.000000Z
Interviewing Ross Taylor on the state of AI: Chinese open models, scaling reasoning, useful tools, and what comes next
Interconnects 2025-09-25T10:01:35.000000Z
Yann LeCun团队新作LLM-JEPA：结合联合嵌入预测架构，显著提升大模型微调性能与效率，在代码生成任务上表现卓越
我爱计算机视觉 2025-09-23T15:08:30.000000Z
RLHF要下岗？Meta × 牛津搞出新套路：用算力教算力，大模型训练新范式来了！
PaperWeekly 2025-09-22T16:06:47.000000Z
大模型训练新突破！Meta提出LSP：无数据也能实现能力飞升
新智元 2025-09-20T11:35:56.000000Z
大模型训练新突破！Meta提出LSP：无数据也能实现能力飞升
新智元 2025-09-20T10:33:17.000000Z
A key type of AI training data is running out. Googlers have a bold new idea to fix that.
All Content from Business Insider 2025-09-15T16:45:10.000000Z
Meta超级智能实验室新论文陷争议！被指忽略大量前人研究
智源社区 2025-09-13T08:38:34.000000Z
大模型“记性差一点”反而更聪明，金鱼损失随机剔除token，让AI不再死记硬背
36kr 2025-09-03T23:55:54.000000Z