Daily Arxiv

Esta página recopila y organiza artículos sobre inteligencia artificial publicados en todo el mundo.
La información aquí presentada se resume utilizando Google Gemini y el sitio se gestiona sin fines de lucro.
Los derechos de autor de los artículos pertenecen a sus autores y a las instituciones correspondientes; al compartir el contenido, basta con citar la fuente.

Modelos de lenguaje de la Fundación Apple Intelligence: Informe tecnológico 2025

Created by
  • Haebom

Autor

Ethan Li (Taoyi), Anders Boesen Lindbo Larsen (Taoyi), Chen Zhang (Taoyi), Xiyou Zhou (Taoyi), Jun Qin (Taoyi), Dian Ang Yap (Taoyi), Narendran Raghavan (Taoyi), Xuankai Chang (Taoyi), Margit Bowler (Taoyi), Eray Yildiz (Taoyi), John Peebles (Taoyi), Hannah Gillis Coleman (Taoyi), Matteo Ronchi (Taoyi), Peter Gray (Taoyi), Keen You (Taoyi), Anthony Spalvieri-Kruse (Taoyi), Ruoming Pang (Taoyi), Reed Li (Taoyi), Yuli Yang (Taoyi), Emad Soroush (Taoyi), Zhiyun Lu (Taoyi), Crystal Xiao (Taoyi), Rong Situ (Taoyi), Jordan Huffaker (Taoyi), David Griffiths (Taoyi), Zaid Ahmed (Taoyi), Peng Zhang (Taoyi), Daniel Parilla (Taoyi), Asaf Liberman (Taoyi), Jennifer Mallalieu (Taoyi), Parsa Mazaheri (Taoyi), Qibin Chen (Taoyi), Manjot Bilkhu (Taoyi), Aonan Zhang (Taoyi), Eric Wang (Taoyi), Dave Nelson (Taoyi), Michael FitzMaurice (Taoyi), Thomas Voice (Taoyi), Jeremy Liu (Taoyi), Josh Shaffer (Taoyi), Shiwen Zhao (Taoyi), Prasanth Yadla (Taoyi), Farzin Rasteh (Taoyi), Pengsheng Guo (Taoyi), Arsalan Farooq (Taoyi), Jeremy Snow (Taoyi), Stephen Murphy (Taoyi), Tao Lei (Taoyi), Minsik Cho (Taoyi), George Horrell (Taoyi), Sam Dodge (Taoyi), Lindsay Hislop (Taoyi), Sumeet Singh (Taoyi), Alex Dombrowski (Taoyi), Aiswarya Raghavan (Taoyi), Sasha Sirovica (Taoyi), Mandana Saebi (Taoyi), Faye Lao (Taoyi), Max Lam (Taoyi), TJ Lu (Taoyi), Zhaoyang Xu (Taoyi), Karanjeet Singh (Taoyi), Marc Kirchner (Taoyi), David Mizrahi (Taoyi), Rajat Arora (Taoyi), Haotian Zhang (Taoyi), Henry Mason (Taoyi), Lawrence Zhou (Taoyi), Yi Hua (Taoyi), Ankur Jain (Taoyi), Felix Bai (Taoyi), Joseph Astrauskas (Taoyi), Floris Weers (Taoyi), Josh Gardner (Taoyi), Mira Chiang (Taoyi), Yi Zhang (Taoyi), Pulkit Agrawal (Taoyi), Tony Sun (Taoyi), Quentin Keunebroek (Taoyi), Matthew Hopkins (Taoyi), Bugu Wu (Taoyi), Tao Jia (Taoyi), Chen Chen (Taoyi), Zirui Wang (Taoyi), Cha Chen (Taoyi), Xiaoyi Ren (Taoyi), Feng Nan (Taoyi), Josh Elman (Taoyi), Dong Yin (Taoyi), Yusuf Goren (Taoyi), Jeff Lai (Taoyi), Yiran Fei (Taoyi), Syd Evans (Taoyi), Muyang Yu (Taoyi), Guoli Yin (Taoyi), Yi Qin (Taoyi), Erin Feldman (Taoyi), Isha Garg (Taoyi), Aparna Rajamani (Taoyi), Karla Vega (Taoyi), Walker Cheng (Taoyi), TJ Collins (Taoyi), Hans Han (Taoyi), Raul Rea Menacho (Taoyi), Simon Yeung (Taoyi), Sophy Lee (Taoyi), Phani Mutyala (Taoyi), Ying-Chang Cheng (Taoyi), Zhe Gan (Taoyi), Sprite Chu (Taoyi), Justin Lazarow (Taoyi), Alessandro Pappalardo (Taoyi), Federico Scozzafava (Taoyi), Jing Lu (Taoyi), Erik Daxberger (Taoyi), Laurent Duchesne (Taoyi), Jen Liu (Taoyi), David Guera (Taoyi), Stefano Ligas (Taoyi), Mary Beth Kery (Taoyi), Brent Ramerth (Taoyi), Ciro Sannino (Taoyi), Marcin Eichner (Taoyi), Haoshuo Huang (Taoyi), Rui Qian (Taoyi), Moritz Schwarzer-Becker (Taoyi), David Riazati (Taoyi), Mingfei Gao (Taoyi), Bailin Wang (Taoyi), Jack Cackler (Taoyi), Yang Lu (Taoyi), Ransen Niu (Taoyi), John Dennison (Taoyi), Guillaume Klein (Taoyi), Jeffrey Bigham (Taoyi), Deepak Gopinath (Taoyi), Navid Shiee (Taoyi), Darren Botten (Taoyi), Guillaume Tartavel (Taoyi), Alex Guillén García (Taoyi), Sam Xu (Taoyi), Victoria MonchJuan Haladjian (Taoyi), Zi-Yi Dou (Taoyi), Matthias Paulik (Taoyi), Adolfo López Méndez (Taoyi), Zhen Li (Taoyi), Hong-You Chen (Taoyi), Chao Jia (Taoyi), Dhaval Doshi (Taoyi), Zhengdong Zhang (Taoyi), Raunak Manjani (Taoyi), Aaron Franklin (Taoyi), Zhile Ren (Taoyi), David Chen (Taoyi), Artsiom Peshko (Taoyi), Nandhitha Raghuram (Taoyi), Hans Hao (Taoyi), Jiulong Shan (Taoyi), Kavya Nerella (Taoyi), Ramsey Tantawi (Taoyi), Vivek Kumar (Taoyi), Saiwen Wang (Taoyi), Brycen Wershing (Taoyi), Bhuwan Dhingra (Taoyi), Dhruti Shah (Taoyi), Ob Adaranijo (Taoyi), Xin Zheng (Taoyi), Tait Madsen (Taoyi), Hadas Kotek (Taoyi), Chang Liu (Taoyi), Yin Xia (Taoyi), Hanli Li (Taoyi), Suma Jayaram (Taoyi), Yanchao Sun (Taoyi), Ahmed Fakhry (Taoyi), Vasileios Saveris (Taoyi), Dustin Withers (Taoyi), Yanghao Li (Taoyi), Alp Aygar (Taoyi), Andres Romero Mier Y Teran (Taoyi), Kaiwei Huang (Taoyi), Mark Lee (Taoyi), Xiujun Li (Taoyi), Yuhong Li (Taoyi), Tyler Johnson (Taoyi), Jay Tang (Taoyi), Joseph Yitan Cheng (Taoyi), Futang Peng (Taoyi), Andrew Walkingshaw (Taoyi), Lucas Guibert (Taoyi), Abhishek Sharma (Taoyi), Cheng Shen (Taoyi), Piotr Maj (Taoyi), Yasutaka Tanaka (Taoyi), You-Cyuan Jhang (Taoyi), Vivian Ma (Taoyi), Tommi Vehvilainen (Taoyi), Kelvin Zou (Taoyi), Jeff Nichols (Taoyi), Matthew Lei (Taoyi), David Qiu (Taoyi), Yihao Qian (Taoyi), Gokul Santhanam (Taoyi), Wentao Wu (Taoyi), Yena Han (Taoyi), Dominik Moritz (Taoyi), Haijing Fu (Taoyi), Mingze Xu (Taoyi), Vivek Rathod (Taoyi), Jian Liu (Taoyi), Louis D'hauwe (Taoyi), Qin Ba (Taoyi), Haitian Sun (Taoyi), Haoran Yan (Taoyi), Philipp Dufter (Taoyi), Anh Nguyen (Taoyi), Yihao Feng (Taoyi), Emma Wang (Taoyi), Keyu He (Taoyi), Rahul Nair (Taoyi), Sanskruti Shah (Taoyi), Jiarui Lu (Taoyi), Patrick Sonnenberg (Taoyi), Jeremy Warner (Taoyi), Yuanzhi Li (Taoyi), Bowen Pan (Taoyi), Ziyi Zhong (Taoyi), Joe Zhou (Taoyi), Sam Davarnia (Taoyi), Olli Saarikivi (Taoyi), Irina Belousova (Taoyi), Rachel Burger (Taoyi), Shang-Chen Wu (Taoyi), Di Feng (Taoyi), Bas Straathof (Taoyi), James Chou (Taoyi), Yuanyang Zhang (Taoyi), Marco Zuliani (Taoyi), Eduardo Jimenez (Taoyi), Abhishek Sundararajan (Taoyi), Xianzhi Du (Taoyi), Chang Lan (Taoyi), Nilesh Shahdadpuri (Taoyi), Peter Grasch (Taoyi), Sergiu Sima (Taoyi), Josh Newnham (Taoyi), Varsha Paidi (Taoyi), Jianyu Wang (Taoyi), Kaelen Haag (Taoyi), Alex Braunstein (Taoyi), Daniele Molinari (Taoyi), Richard Wei (Taoyi), Brenda Yang (Taoyi), Nicholas Lusskin (Taoyi), Joanna Arreaza-Taylor (Taoyi), Meng Cao (Taoyi), Nicholas Seidl (Taoyi), Simon Wang (Taoyi), Jiaming Hu (Taoyi), Yiping Ma (Taoyi), Mengyu Li (Taoyi), Kieran Liu (Taoyi), Hang Su (Taoyi), Sachin Ravi (Taoyi), Chong Wang (Taoyi), (Taoyi), Jane Li (Taoyi), Andrew Hansen (Taoyi), Javier Movellan (Taoyi), Christopher Neubauer (Taoyi), Hanzhi Zhou (Taoyi), Chris Chaney (Taoyi), Nazir Kamaldin (Taoyi), Valentin Wolf (Taoyi), Fernando Bermúdez -Medina (Taoyi), Joris Pelemans (Taoyi), Peter Fu (Taoyi), Howard Xing (Taoyi), Xiang Kong (Taoyi), Wayne Shan (Taoyi), Gabriel Jacoby-Cooper (Taoyi), Dongcai Shen (Taoyi), Tom Gunter (Taoyi), Guillaume Seguin (Taoyi), Fangping Shi (Taoyi), Shiyu Li (Taoyi), Yang Xu (Taoyi), Areeba Kamal (Taoyi), Dan Masi (Taoyi), Saptarshi Guha (Taoyi), Qi Zhu (Taoyi), Jenna Thibodeau (Taoyi), Changyuan Zhang (Taoyi), Rebecca Callahan (Taoyi), Charles Maalouf (Taoyi), Wilson Tsao (Taoyi), Boyue Li (Taoyi), Qingqing Cao (Taoyi), Naomy Sabo (Taoyi), Cheng Leong (Taoyi), Yi Wang (Taoyi), Anupama Mann Anupama (Taoyi), Colorado Reed (Taoyi), Kenneth Jung (Taoyi), Zhifeng Chen (Taoyi), Mohana Prasad Sathya Moorthy (Taoyi), Yifei He (Taoyi), Erik Hornberger (Taoyi), Devi Krishna (Taoyi), Senyu Tong (Taoyi), Michael (Taoyi), Lee, David Haldimann, Yang Zhao, Bowen Zhang, Chang Gao, Chris Bartels, Sushma Rao, Nathalie Tran, Simon Lehnerer, Co Giang, Patrick Dong, Junting Pan, Biyao Wang, Dongxu Li, Mehrdad Farajtabar, Dongseong Hwang, Grace Duanmu, Eshan Verma, Sujeeth Reddy, Qi Shan, Hongbin Gao, Nan Du, Pragnya Sridhar, Forrest Huang, Yingbo Wang, Nikhil Bhendawade, Diane Zhu, Sai Aitharaju, Fred Hohman, Lauren Gardiner, Chung-Cheng Chiu, Yinfei Yang, Alper Kokmen, Frank Chu, Ke Ye, Kaan Elgin, Oron Levy, John Park, Donald Zhang, Eldon Schoop, Nina Wenzel, Michael Booker, Hyunjik Kim, Chinguun Erdenebileg, Nan Dun, Eric Liang Yang, Priyal Chhatrapati, Vishaal Mahtani, Haiming Gang, Kohen Chia, Deepa Seshadri, Donghan Yu, Yan Meng, Kelsey Peterson, Zhen Yang, Yongqiang Wang, Carina Peng, Doug Kang, Anuva Agarwal, Albert Antony, Juan Lao Tebar, Albin Madappally Jose, Regan Poston, Andy De Wang, Gerard Casamayor, Elmira Amirloo, Violet Yao, Wojciech Kryscinski, Kun Duan, Lezhi L

Describir

Apple presenta dos modelos de lenguaje multilingües y multimodales que potencian las funciones de Apple Intelligence en sus dispositivos y servicios. Uno es un modelo integrado en el dispositivo con 3000 millones de parámetros, optimizado para Apple Silicon mediante innovaciones arquitectónicas (uso compartido de caché KV y entrenamiento con capacidad de cuantificación de 2 bits). El otro es un modelo de servidor escalable basado en un novedoso transformador de mezcla de pistas paralelas de expertos (PT-MoE) que combina paralelismo de pistas, computación dispersa con expertos y atención global-local intercalada. Ambos modelos se entrenan con conjuntos de datos multilingües y multimodales a gran escala derivados de un rastreo web responsable, corpus con licencia y datos sintéticos de alta calidad, y se mejoran aún más mediante un ajuste fino supervisado y aprendizaje de refuerzo en una nueva plataforma asíncrona. Los modelos resultantes comprenden imágenes, ejecutan llamadas a herramientas y admiten varios idiomas adicionales. Tanto el modelo de servidor como el integrado en el dispositivo tienen un rendimiento similar o incluso superior a las líneas base abiertas de tamaño similar en evaluaciones comparativas públicas y humanas. Un nuevo marco de modelo fundamental centrado en Swift proporciona generación de guías, llamadas a herramientas restringidas y ajuste del adaptador LoRA, lo que permite a los desarrolladores integrar estas funciones con tan solo unas pocas líneas de código. Los últimos avances en los modelos de inteligencia de Apple se basan en el compromiso de Apple con una IA responsable, con innovaciones como la computación en la nube privada y medidas de seguridad como el filtrado de contenido y las clasificaciones regionales, todo ello basado en el compromiso de Apple con la privacidad del usuario.

Takeaways, Limitations

•
Takeaways:
◦
Desarrollo de un modelo de lenguaje multimodal, multilingüe y de alto rendimiento para su aplicación en dispositivos y servicios de Apple.
◦
Se presentan innovaciones arquitectónicas para la optimización de modelos en el dispositivo (uso compartido de caché KV, aprendizaje consciente de la cuantificación de 2 bits).
◦
Arquitectura de transformador PT-MoE propuesta para modelos de servidor escalables.
◦
Enfoques de IA responsables y protecciones mejoradas de la privacidad.
◦
Accesibilidad mejorada para desarrolladores a través de marcos basados ​​en Swift.
◦
Demostró desempeño competitivo en puntos de referencia públicos y evaluaciones humanas.
•
Limitations:
◦
Falta de información detallada sobre el tamaño o la composición del conjunto de datos específicos en el artículo.
◦
Falta de explicación técnica específica sobre la computación en la nube privada.
◦
Faltan detalles sobre un enfoque responsable de la IA. Se necesitan explicaciones detalladas sobre las salvaguardias y los métodos de evaluación específicos.
◦
Falta de consideración del impacto ambiental del modelo (consumo energético, etc.).
◦
Falta de mención de planes de mantenimiento y actualización a largo plazo.
👍
Made with Slashpage