Apple brings new work in powerful ways to the annual natural language conference (EMNLP), which takes place in person from November 4 – 9, in Suzhou, China. EMNLP focuses on research surrounding the science and technology of spoken language processing.
Below is the schedule of Apple-sponsored lectures and events at EMNLP 2025.
Skip to section:
The program
Stop by the Apple Booth at the Suzhou International Expo Center during the show. All times listed in CST (SUZHOU local time):
- Wednesday, November 5: 09:00 – 18:00
- Thursday, November 6: 09:00 – 18:00
- Friday, November 7: 09:00 – 16:00
The program
Wednesday, November 5
- The situation
- The MLX
- 09:00 – 18:00The apple booth
- We demonstrate extensive modeling and training on a device using MLX. MLX is a dynamic simulation system developed by Apple Silicon and delivered by Apple Machine Learning Research. It enables the training and subjection of various altered models on apple silicon powered devices with greater intelligence and flexibility. In this demo we show a good configuration of the 7b parameter of LLM on iPhone, Image generation using Effance large model on iPad and text generation using large language models in a cluster of four studios.
- Decorative photo paper
- Towards machine translation: Studies in human translation studies
- 11:00 – 12:30Poster Session 1, Hall C3
- Mathias Sperber, Maureen de Seyssel, Jayson Bao, Matthias Paulik
- Presentation, poster
- Testing the metrics of the test – the miracle of hallucination detection
- 13:00 – 14:00Findings 1, Hall C3
- Atharva Kulkarni (USC), Yuan Zhang, Joyash Ruben Antony Moniz (Doordash), Hugh G GE, Dhiv Geng, Swabha Swayamdipta (USC), Hong Yu
Thursday, November 6
- The situation
- The MLX
- 9:00 – 18:00The apple booth
- We demonstrate extensive modeling and training on a device using MLX. MLX is a dynamic simulation system developed by Apple Silicon and delivered by Apple Machine Learning Research. It enables the training and subjection of various altered models on apple silicon powered devices with greater intelligence and flexibility. In this demo we show a good configuration of the 7b parameter of LLM on iPhone, Image generation using Effance large model on iPad and text generation using large language models in a cluster of four studios.
Friday, November 7
- Workhouse
- Polynorm: Shot-based normalized text-to-speech vlm
- 08:00 – 09:00Recruit session 4 (Virtual)
- Michel Wong, Haotian He, Ali Alshehri, Sophia Kao
Saturday, November 8
Papers accepted
WritersEileen Pan †, Anna Seo Gyeong Choi †, Maartje Ter Hoeve, Skyler Seto, Allison Koenecke †
Exploring the Role of Data Quality in Training Linguistic Models of the Zulu Language
Skyler Seto, Maartje Ter Hoeve, Maureen de Seyssel, David Grangier
WritersNivhie Sivakumar *, Natalie Mackraz *, Krishna Khorshidi, Krishna Patel †, Barry-John Theobald, Luca Zappella, Nicholas Artoroff
WritersXinze Wang, Chen Chen, Yinfei Yang, Hong-You Chen, Bowen Zhang, Aditya Pal, Xiangxin Zhu, Xianzhin zhu
WritersMaureen de Seyssel *, Jie Chi *, Skyler Seto, Maartje Ter Hoeve, Masha Fedzechkina, Natalie Schleter
WritersAtharva Kulkarni † *, yuan zhang, Joel ruben antony moniz *, xiu ge, ge, hsiang tseng, dhivya piraviperumal, swabha swayamdipta †, Hong Yu
WritersBrihi Joshi ** †, Xiang Ren †, Swabha Swayamdipta †, Rik kolog-kedziorski, Tim Paek
WritersYajie Li †, Albert Gatta Genapani †, Pujitha Thejaswi †, De Meng, Priyanshu Kumar, Saloni Poutdar
WritersSong of Congzheng, Xinya Tang
Polynorm: Shot-based normalized text-to-speech vlm
Michel Wong, Haotian He, Ali Alshehri, Sophia Kao
WritersRik Koncel-Kedziorski, Brihi Joshi †, Tim Paek
WritersNikhil Bhendawade, Irina Belousova, Qichen Fu, Henry Mason, Mohammad Rastegari, Mahyar Najibi
Towards machine translation: Studies in human translation studies
Mathias Sperber, Maureen de Seyssel, Jayson Bao, Matthias Paulik
Acceptance
Saloni Potdar is the seat of the sector track of EMNLP 2025.
QingQing Cao and Natalie Scluter are the local high chairs.
Richard Bai, Maartje Ter Hoeve, Chao Jiang, Rik kologyski, and Yzhe Zhang are local chairs.
Jason Dong, Lu Ren and Shu W. are session chairs.
Richard Bai is a meeting to work for the meeting of Extending natural language processing (WinLP).
Maartje Ter Hoeve, Katherine MetarCalf, and Andrew Silva are the workshop organizers Tailoging AI: Assessing Practical Practice and Performance LLM Fellow (Pals).
Natalie Schuterute and Barry-John Theobald serve on the workshop's advisory board Tailoging AI: Assessing Practical Practice and Performance LLM Fellow (Pals).
Maureen de Seysser is a workshop reviewer for Tailoging AI: Assessing Practical Practice and Performance LLM Fellow (Pals).
Aswarth Abhilash Dara, Mozhdeh Gheini, and Stephan Peitz are reviewers for the EMNLP 2025 industry track.
Jie Chi, Jason Dong, qin Gao, Xintong Geng, Alex Ppangelis, Shuwen Qiu, Robin qiu, and Robin qu, and Hong Yu are EMNLP 2025 reviewers.
