[LG]《Tree-OPO: Off-policy Monte Carlo Tr
[LG]《Tree-OPO: Off-policy Monte Carlo Tr
爱生活爱珂珂的文章
![[LG]《Tree-OPO: Off-policy Monte Carlo Tr](http://dayu-img.uc.cn/columbus/img/oc/1002/3a5232d8d3c8719fa71438b30a167993.jpg)
![[LG]《Sharing is Caring: Efficient LM Pos](http://dayu-img.uc.cn/columbus/img/oc/1002/655a8884f8390e01ce55236243af1a42.jpg)
[LG]《Sharing is Caring: Efficient LM Pos
[LG]《Sharing is Caring: Efficient LM Pos
![[CL]《Saturation-Driven Dataset Generatio](http://dayu-img.uc.cn/columbus/img/oc/1002/d08388d845de91318b522b23f4de1b08.jpg)
[CL]《Saturation-Driven Dataset Generatio
[CL]《Saturation-Driven Dataset Generatio
![早！[太阳] 早安 ](http://dayu-img.uc.cn/columbus/img/oc/1002/696cd70b3ca74819849709e0d266ca43.jpg)
早！[太阳] 早安
早！[太阳] 早安
《Send to which account? Evaluation of an
《Send to which account? Evaluation of an
《Jupiter: Enhancing LLM Data Analysis Ca
《Jupiter: Enhancing LLM Data Analysis Ca
![[CL]《Large Language Model Hacking: Quant](http://dayu-img.uc.cn/columbus/img/oc/1002/a08c03c84cda61096ebf07c954744467.jpg)
[CL]《Large Language Model Hacking: Quant
[CL]《Large Language Model Hacking: Quant
![[CL]《mmBERT: A Modern Multilingual Encod](http://dayu-img.uc.cn/columbus/img/oc/1002/0b830193d279883bc099996fcba3b66c.jpg)
[CL]《mmBERT: A Modern Multilingual Encod
[CL]《mmBERT: A Modern Multilingual Encod
![[LG]《Probing the Preferences of a Langua](http://dayu-img.uc.cn/columbus/img/oc/1002/f715c000df5c034a7b30bfa34dae91b5.jpg)
[LG]《Probing the Preferences of a Langua
[LG]《Probing the Preferences of a Langua
![[LG]《The Illusion of Diminishing Returns](http://dayu-img.uc.cn/columbus/img/oc/1002/6065389338601f6a31f228c4390b0cbc.jpg)
[LG]《The Illusion of Diminishing Returns
[LG]《The Illusion of Diminishing Returns
![[LG]《Reverse-Engineered Reasoning for Op](http://dayu-img.uc.cn/columbus/img/oc/1002/6460676a022f34bcac22327cc9aace5b.jpg)
[LG]《Reverse-Engineered Reasoning for Op
[LG]《Reverse-Engineered Reasoning for Op
![早！[太阳] 早安 ](http://dayu-img.uc.cn/columbus/img/oc/1002/fe40d9a0a9d3524949d5a8e23e82581b.jpg)
早！[太阳] 早安
早！[太阳] 早安
![晚安～ [月亮] 晚安 ](http://dayu-img.uc.cn/columbus/img/oc/1002/12aa2a1f4f4f8a2052ba9de0bec7b78a.jpg)
晚安～ [月亮] 晚安
晚安～ [月亮] 晚安
《爱可可微博热门分享(9.13)》
《爱可可微博热门分享(9.13)》
《Attention sinks from the graph perspect
《Attention sinks from the graph perspect
一套能将任何模糊想法转化为精确、高效 AI 指令的系统化框架——Lyra 提示词
一套能将任何模糊想法转化为精确、高效 AI 指令的系统化框架——Lyra 提示词
![[人人能懂] 回音室、水印破解与选择之道本期我们要聊一个有点令人不安的现实：当A](http://dayu-img.uc.cn/columbus/img/oc/1002/2b2bf59295ec0dcb2d539cff4f5c6596.jpg)
[人人能懂] 回音室、水印破解与选择之道本期我们要聊一个有点令人不安的现实：当A
[人人能懂] 回音室、水印破解与选择之道本期我们要聊一个有点令人不安的现实：当A
《Writing effective tools for agents — wi
《Writing effective tools for agents — wi
![[CL]《CDE: Curiosity-Driven Exploration f](http://dayu-img.uc.cn/columbus/img/oc/1002/862e4b254142c5883cee25d317ad299c.jpg)
[CL]《CDE: Curiosity-Driven Exploration f
[CL]《CDE: Curiosity-Driven Exploration f
![[LG]《An entropy formula for the Deep Lin](http://dayu-img.uc.cn/columbus/img/oc/1002/049167b0643151960e696b5b8c73510c.jpg)
[LG]《An entropy formula for the Deep Lin
[LG]《An entropy formula for the Deep Lin
作者信息
