論文標題：Direct Preference Optimization: Your Language Model is Secretly a R...
投稿
收錄了118篇文章 · 27人關(guān)注
-
直接偏好優(yōu)化技術(shù)DPO基礎(chǔ)理論及推導
-
自適應視圖增強的謠言檢測圖對比學習方法論文標題：Propagation Tree Is Not Deep: Adaptive Graph Contrastive Learning A...
-
生成式大模型的RLHF技術(shù)（一）：基礎(chǔ)一、概述 大語言模型（LLMs）在預訓練的過程中通常會捕捉數(shù)據(jù)的特征额港，而這些訓練數(shù)據(jù)通常既包含高質(zhì)量的也包含低質(zhì)量的蒿涎，因此模型有時會產(chǎn)生不被期望...
-
LoRA：大模型下游任務的低秩適應論文標題：LoRA: Low-Rank Adaptation of Large Language Models論文鏈接：https://arxi...
-
Megatron-LM：Transformer模型專用分布式張量模型并行方法論文標題：Megatron-LM: Training Multi-Billion Parameter Language Models Using...
-
思維樹：大模型的復雜推理技術(shù)論文標題：Tree of Thoughts: Deliberate Problem Solving with Large Language Mo...
-
LIMA：小規(guī)模監(jiān)督數(shù)據(jù)指令微調(diào)論文標題：LIMA: Less Is More for Alignment論文鏈接：https://arxiv.org/abs/2305.112...
-
語言模型的自洽性思維鏈推理技術(shù)論文標題：Self-Consistency Improves Chain of Thought Reasoning in Language Mo...
-
GPipe：微批量流水線并行論文標題：GPipe: Easy Scaling with Micro-Batch Pipeline Parallelism論文鏈接：https...
-
InstructGPT：語言模型的人類反饋指令對齊
論文標題：Training language models to follow instructions with human feedback...
{var%20f='http://v.t.sina.com.cn/share/share.php?appkey=1881139527',u=z||d.location,p=['&url=',e(u),'&title=',e(t||d.title),'&source=',e(r),'&sourceUrl=',e(l),'&content=',c||'gb2312','&pic=',e(p||'')].join('');function%20a(){if(!window.open([f,p].join(''),'mb',['toolbar=0,status=0,resizable=1,width=440,height=430,left=',(s.width-440)/2,',top=',(s.height-430)/2].join('')))u.href=[f,p].join('');};if(/Firefox/.test(navigator.userAgent))setTimeout(a,0);else%20a();})(screen,document,encodeURIComponent,'','','https://upload.jianshu.io/collections/images/1904333/%E7%BC%96%E7%A8%8B2.jpg', '推薦專題《人工智能》（ 分享自 @簡書 ）','http://www.reibang.com/c/90f5201e753b?utm_campaign=maleskine&utm_content=collection&utm_medium=reader_share&utm_source=weibo','頁面編碼gb2312|utf-8默認gb2312'));){kind=link}