Apple is presenting new research at the International Conference on Machine Learning (ICML 2025), which takes place in person in Vancouver, Canada, from July 13 - 19. We are proud to again sponsor the conference, which focuses on presenting and publishing cutting-edge research in machine learning, artificial intelligence, statistics, data science, and related application areas like machine vision, computational biology, and robotics. Below is an overview of Apple's participation at ICML 2025.
Jump to a section:
Schedule
Stop by the Apple booth (#307) in the Vancouver Convention Center during exhibition hours. All times listed in PT (Vancouver time):
- Monday, July 14: 4:00 PM - 8:00 PM
- Tuesday, July 15: 9:30 AM - 6:00 PM
- Wednesday, July 16: 9:30 AM - 6:00 PM
- Thursday, July 17: 9:30 AM - 1:00 PM
Sunday, July 13
- EXPO TALK
- "Distillation Scaling Laws"
- 5:00pm - 6:00pm
West Ballroom A - Dan Busbridge will be giving the Apple Expo Talk.
Monday, July 14
- AFFINITY GROUP
- LatinX in AI Workshop
- 3:00pm - 4:00pm
West Meeting Room 118-120 - Lauren Araujo and Miguel Sarabia del Castillo will represent Apple at the LXAI mentor panel discussion.
Tuesday, July 15
- POSTER
- "Cavia: Camera-controllable Multi-view Video Diffusion with View-Integrated Attention"
- 11:00am - 1:30pm
Poster Session 1 West, #W-207, West Exhibition Hall B2-B3 - Dejia Xu (UT Austin), Yifan Jiang, Chen (Kimi) Huang, Liangchen Song, Thorsten Gernoth, Liangliang Cao (Google), Atlas Wang (UT Austin), Hao Tang
- POSTER
- "CtrlSynth: Controllable Image Text Synthesis for Data-Efficient Multimodal Learning"
- 11:00am - 1:30pm
Poster Session 1 East, #E-2103, East Exhibition Hall A-B - Qingqing Cao, Mahyar Najibi, Sachin Mehta (Meta [work done while at Apple])
- POSTER
- "Local Pan-privacy for Federated Analytics"
- 11:00am - 1:30pm
Poster Session 1 East, #E-903, East Exhibition Hall A-B - Vitaly Feldman, Audra McMillan, Guy Rothblum, Kunal Talwar
- POSTER
- "Soup of Experts: Pretraining Domain-Adaptive Models via Weights Averaging"
- 11:00am - 1:30pm
Poster Session 1 East, #E-1702, East Exhibition Hall A-B - Pierre Ablin, Angelos Katharopoulos, Skyler Seto, David Grangier
- POSTER
- "Tracking the Best Expert Privately"
- 11:00am - 1:30pm
Poster Session 1 East, #E-1008, East Exhibition Hall A-B - Hilal Asi, Aadi Saha, Vinod Raman (University of Michigan)
- ORAL
- "Addressing Misspecification in Simulation-based Inference through Data-driven Calibration"
- 4:15pm - 4:30pm
West Ballroom D, Oral 2E Optimal Transport - Antoine Wehenkel, Juan L. Gamella, Ozan Sener, Jens Behrmann, Guillermo Sapiro, Jörn-Henrik Jacobsen, Marco Cuturi
- SPOTLIGHT POSTER
- "Addressing Misspecification in Simulation-based Inference through Data-driven Calibration"
- 4:30pm - 7:00pm
Poster Session 2 East, #E-1404, East Exhibition Hall A-B - Antoine Wehenkel, Juan L. Gamella, Ozan Sener, Jens Behrmann, Guillermo Sapiro, Jörn-Henrik Jacobsen, Marco Cuturi
- POSTER
- "CommVQ: Commutative Vector Quantization for KV Cache Compression"
- 4:30pm - 7:00pm
Poster Session 2 East, #E-2711, East Exhibition Hall A-B - Junyan Li (University of Massachusetts Amherst), Yang Zhang (Massachusetts Institute of Technology), Muhammad Yusuf Hassan (University of Massachusetts Amherst), Talha Chafekar (University of Massachusetts Amherst), Tianle Cai (Princeton University), Zhile Ren, Pengsheng Guo, Binazir Karimzadeh, Colorado Reed, Chong Wang, Chuang Gan (University of Massachusetts Amherst)
- POSTER
- "Shielded Diffusion: Generating Novel and Diverse Images using Sparse Repellency"
- 4:30pm - 7:00pm
Poster Session 2 East, #E-3103 - Michael Kirchhof, James Thornton, Pierre Ablin, Louis Béthune, Michal Klein, Eugene Ndiaye, Marco Cuturi
- POSTER
- "FlexTok: Resampling Images into 1D Token Sequences of Flexible Length"
- 4:30pm - 7:00pm
Poster Session 2 East, #E-3010, East Exhibition Hall A-B - Roman Bachmann, Jesse Allardice, David Mizrahi, Enrico Fini, Oğuzhan Fatih Kar (EPFL), Elmira Amirloo, Alaaeldin El-Nouby, Amir Zamir (EPFL), Afshin Dehghan
Wednesday, July 16
- POSTER
- "Scaling Laws for Forgetting during Finetuning with Pretraining Data Injection"
- 11:00am - 1:30pm
Poster Session 2 East, #E-2708, East Exhibition Hall A-B - Louis Béthune, David Grangier, Dan Busbridge, Eleonora Gualdoni, Marco Cuturi, Pierre Ablin
- POSTER
- "Distillation Scaling Laws"
- 11:00am - 1:30pm
Poster Session 3 East, #E-2310, East Exhibition Hall A-B - Dan Busbridge, Jason Ramapuram, Floris Weers, Amitis Shidani (University of Oxford), Russ Webb, Etai Littwin
- POSTER
- "Beyond Sensor Data: Foundation Models of Behavioral Data from Wearables Improve Health Predictions"
- 11:00am - 1:30pm
Poster Session 3 West, #W-302, West Exhibition Hall B2-B3 - Eray Erturk (USC), Fahad Kamran, Salar Abbaspourazad, Sean Jewell, Harsh Sharma, Nick Foti, Joe Futoma
- POSTER
- "INRFlow: Flow Matching for INRs in Ambient Space"
- 11:00am - 1:30pm
Poster Session 3 East, #E-3407, East Exhibition Hall A-B - Yuyang Wang, Anurag Ranjan, Josh Susskind, Miguel Angel Bautista Martin
- POSTER
- "Instruction-Following Pruning for Large Language Models"
- 11:00am - 1:30pm
Poster Session 3 East, #E-2711, East Exhibition Hall A-B - Bairu Hou (UCSB), Qibin Chen, Jianyu Wang, Guoli Yin, Chong Wang, Nan Du, Ruoming Pang, Shiyu Chang (UCSB), Tao Lei
- POSTER
- "Is Your Model Fairly Certain? A Fine-grained Approach Towards LLM Fairness"
- 11:00am - 1:30pm
Poster Session 3 East, #E-905, East Exhibition Hall A-B - Oliver Wang (Carnegie Mellon University), Niv Sivakumar, Falaah Arif Khan (New York University), Rin Metcalf Susa, Adam Golinski, Natalie Mackraz, Barry Theobald, Luca Zappella, Nick Apostoloff
- POSTER
- "Preference Reasoning by Evaluating Decomposed preferences Inferred from Counterfactual Trajectories"
- 11:00am - 1:30pm
Poster Session 3 East, #E-2704, East Exhibition Hall A-B - Stéphane Aroca-Ouellette, Natalie Mackraz, Barry Theobald, Rin Metcalf Susa
- POSTER
- "Proxy-FDA: Proxy-based Feature Distribution Alignment for Fine-tuning Vision Foundation Models without Forgetting"
- 11:00am - 1:30pm
Poster Session 3 East, #E-1309, East Exhibition Hall A-B - Chen Huang, Skyler Seto, Hadi Pour Ansari, Mehrdad Farajtabar, Raviteja Vemulapalli, Fartash Faghri, Oncel Tuzel, Barry Theobald, Josh Susskind
- POSTER
- "Target Concrete Score Matching: A Holistic Framework for Discrete Diffusion"
- 11:00am - 1:30pm
Poster Session 3 East, #E-3310, East Exhibition Hall A-B - Ruixiang Zhang, Shuangfei Zhai, Yizhe Zhang, James Thornton, Zijing Ou (Imperial College London), Josh Susskind, Navdeep Jaitly
- POSTER
- "Parameters vs FLOPs: Scaling Laws for Optimal Sparsity for Mixture-of-Experts Language Models"
- 11:00am - 1:30pm
Poster Session 3 East, #E-2810, East Exhibition Hall A-B - Samira Abnar, Harshay Shah (MIT), Dan Busbridge, Alaaeldin Mohamed Elnouby Ali, Josh Susskind, Vimal Thilak
- POSTER
- "Variational Rectified Flow Matching"
- 11:00am - 1:30pm
Poster Session 3 East, #E-3306, East Exhibition Hall A-B - Pengsheng Guo, Alex Schwing
- AFFINITY GROUP
- Women in Machine Learning (WiML)
- 1:00pm - 2:00pm
West Meeting Room 211-214 - Lucie Perotta, Tatiana Likhomanenko, and Wasifa Jamal will represent Apple at the WiML mentorship roundtable luncheon
- POSTER
- "A Variational Framework for Improving Naturalness in Generative Spoken Language Models"
- 4:30pm - 7:00pm
Poster Session 4 West, #W-411, West Exhibition Hall B2-B3 - Li-Wei Chen (Carnegie Mellon University), Takuya Higuchi, Zak Aldeneh, Ahmed Hussen Abdelaziz, Alexander Rudnicky (Carnegie Mellon University)
- POSTER
- "Learning to Route LLMs with Confidence Tokens"
- 4:30pm - 7:00pm
Poster Session 4 East, #E-1208, East Exhibition Hall A-B - Allen Chuang (Rice University), Prathusha Kameswara Sarma, Parikshit Gopalan, John Boccio, Sara Bolouki, Xia (Ben) Hu (Rice University), Helen Zhou
- POSTER
- "Mechanisms of Projective Composition of Diffusion Models"
- 4:30pm - 7:00pm
Poster Session 4 East, #E-3105, East Exhibition Hall A-B - Arwen Bradley, Preetum Nakkiran, James Thornton, David Berthelot, Josh Susskind
Thursday, July 17
- POSTER
- "Faster Rates for Private Adversarial Bandits"
- 11:00am - 1:30pm
Poster Session 5 East, #E-905, East Exhibition Hall A-B - Hilal Asi, Vinod Raman (University of Michigan), Kunal Talwar
- POSTER
- "Improved Sample Complexity for Private Nonsmooth Nonconvex Optimization"
- 11:00am - 1:30pm
Poster Session 5 East, #E-901, East Exhibition Hall A-B - Guy Kornowski (Weizmann Institute of Science), Daogao Liu (University of Washington), Kunal Talwar
- POSTER
- "QuantSpec: Self-Speculative Decoding with Hierarchical Quantized KV Cache"
- 11:00am - 1:30pm
Poster Session 5 East, #E-2608, East Exhibition Hall A-B - Rishabh Tiwari (UC Berkeley), Haocheng Xi (UC Berkeley), Aditya Tomar (UC Berkeley), Coleman Hooper (UC Berkeley), Sehoon Kim (UC Berkeley), Max Horton, Mahyar Najibi, Michael W. Mahoney (UC Berkeley), Kurt Keutzer (UC Berkeley), Amir Gholaminejad (UC Berkeley)
- POSTER
- "SPD: Sync-Point Drop for Efficient Tensor Parallelism of Large Language Models"
- 11:00am - 1:30pm
Poster Session 5 West, #W-507, West Exhibition Hall B2-B3 - Han-Byul Kim, Minsik Cho, Duc Hoang, Arnav Kundu, Mohammad Samragh Razlighi
- POSTER
- "Training Software Engineering Agents and Verifiers with SWE-Gym"
- 11:00am - 1:30pm
Poster Session 5 East, #E-2602, East Exhibition Hall A-B - Jiayi Pan (UCB), Xingyao Wang (UIUC), Graham Neubig (CMU), Navdeep Jaitly, Ji Heng (UIUC), Alane Suhr (UCB), Yizhe Zhang
- POSTER
- "Understanding Input Selectivity in Mamba: Impact on Approximation Power, Memorization, and Associative Recall Capacity"
- 11:00am - 1:30pm
Poster Session 5 East, #E-2312, East Exhibition Hall A-B - Teresa Huang (Johns Hopkins University), Abhinav Moudgil (Concordia University, Montréal, Canada), Pau Rodriguez Lopez, Miguel Sarabia del Castillo, Luca Zappella, Federico Danieli
- ORAL
- "Normalizing Flows are Capable Generative Models"
- 4:00pm - 4:15pm
West Ballroom A, Oral 6B Deep Learning Architectures - Shuangfei Zhai, Ruixiang Zhang, Preetum Nakkiran, David Berthelot, Jiatao Gu, Huangjie Zheng, Tianrong Chen, Miguel Angel Bautista Martin, Navdeep Jaitly, Josh Susskind
- POSTER
- "Contrastive Localized Language-Image Pre-Training"
- 4:30pm - 7:00pm
Poster Session 6 East, #E-3209, East Exhibition Hall A-B - Hong-You Chen, Jeff Lai, Haotian Zhang, Angie Wang, Marcin Eichner, Keen You, Meng Cao, Bowen Zhang, Yinfei Yang, Zhe Gan
- POSTER
- "Normalizing Flows are Capable Generative Models"
- 4:30pm - 7:00pm
Poster Session 6 East, #E-2911, East Exhibition Hall A-B - Shuangfei Zhai, Ruixiang Zhang, Preetum Nakkiran, David Berthelot, Jiatao Gu, Huangjie Zheng, Tianrong Chen, Miguel Angel Bautista Martin, Navdeep Jaitly, Josh Susskind
- POSTER
- "Robust Autonomy Emerges from Self-Play"
- 4:30pm - 7:00pm
Poster Session 6 West, #W-821, West Exhibition Hall B2-B3 - Marco Cusumano-Towner, David Hafner, Alex Hertzberg, Brody Huval, Aleksei Petrenko, Eugene Vinitsky, Erik Wijmans, Taylor Killian, Stuart Bowers, Ozan Sener, Philipp Kraehenbuehl, Vladlen Koltun
Friday, July 18
- WORKSHOP PAPER
- "Boolformer: Symbolic Regression of Logic Functions with Transformers"
- 2nd AI for Math Workshop
8:30am - 5:15pm, West Ballroom C - Stephane d'Ascoli (EPFL), Arthur Renard (EPFL), Emmanuel Abbe, Clément Hongler (EPFL), Vassilis Joseph Papadopoulos (EPFL), Josh Susskind, Samy Bengio
- WORKSHOP PAPER
- "Overcoming Vocabulary Constraints with Pixel-level Fallback"
- Tokenization Workshop
8:30am - 5:15pm, West Meeting Room 111-112 - Jonas F. Lotz (University of Copenhagen), Hendra Setiawan, Stephan Peitz, Yova Kementchedjhieva (Mohamed bin Zayed University of Artificial Intelligence)
Saturday, July 19
- WORKSHOP PAPER
- "Evaluating Sample Utility for Data Selection by Mimicking Model Weights"
- DataWorld: Unifying Data Curation Frameworks Across Domains 2025
8:30am - 5:00pm, West Meeting Room 208-209 - Brian Huang (University of Wisconsin-Madison), Manjot Bilkhu, Frederic Sala (University of Wisconsin-Madison), Javier Movellan
- WORKSHOP PAPER
- "Lightweight Safety Guardrails via Synthetic Data and RL-guided Adversarial Training"
- Data in Generative Models Workshop
8:30am - 5:00pm, West Ballroom A - Rachel Li, Riyaaz Shaik, Noyan Tokgozoglu, Aleksei Ilin, Amber Ma, Gor Matevosyan, Suhaa Dada, James Rae
- WORKSHOP PAPER
- "Lightweight Safety Guardrails via Synthetic Data and RL-guided Adversarial Training"
- DataWorld: Unifying Data Curation Frameworks Across Domains 2025
8:30am - 5:00pm, West Meeting Room 208-209 - Rachel Li, Riyaaz Shaik, Noyan Tokgozoglu, Aleksei Ilin, Amber Ma, Gor Matevosyan, Suhaa Dada, James Rae
- WORKSHOP PAPER
- "Lightweight Safety Guardrails via Synthetic Data and RL-guided Adversarial Training"
- Workshop on Reliable and Responsible Foundation Models 2025
8:50am - 6:00pm, West Ballroom C - Rachel Li, Riyaaz Shaik, Noyan Tokgozoglu, Aleksei Ilin, Amber Ma, Gor Matevosyan, Suhaa Dada, James Rae
- WORKSHOP PAPER
- "Self-reflective Uncertainties: Do LLMs Know Their Internal Answer Distribution?"
- Workshop on Reliable and Responsible Foundation Models 2025
8:50am - 6:00pm, West Ballroom C - Michael Kirchhof, Luca Füger (independent researcher), Adam Golinski, Eeshan Gunesh Dhekane, Arno Blaas, Seong Joon Oh (University of Tübingen, Tübingen AI Center), Sinead Williamson
- WORKSHOP PAPER
- "The Geometries of Truth are Orthogonal Across Tasks"
- Workshop on Reliable and Responsible Foundation Models 2025
8:50am - 6:00pm, West Ballroom C - Waïss Azizian, Louis Béthune, Pierre Ablin, Eugene Ndiaye, Michal Klein, Michael Kirchhof, Marco Cuturi
Booth Programming & Demos
Visit Apple's booth (#307) at the Vancouver Convention Center during exhibition hours.
- IN-BOOTH POSTER SESSION
- "Shielded Diffusion: Generating Novel and Diverse Images using Sparse Repellency (SPELL)"
- Tuesday, July 15, 9:30am - 12:00pm
- Presenter: Michael Kirchhof
- IN-BOOTH POSTER SESSION
- "Normalizing Flows are Capable Generative Models"
- Wednesday, July 16, 9:30am - 12:00pm
- Presenter: Jiatao Gu
- IN-BOOTH POSTER SESSION
- "Addressing Misspecification in Simulation-based Inference through Data-driven Calibration"
- Thursday, July 17, 9:30am - 11:30am
- Presenter: Antoine Wehenkel
Technical Demos
- DEMO
- MLX
- MLX is a flexible array framework that is optimized for Apple silicon, and brought to you by Apple Machine Learning Research. It enables training and inference of arbitrarily complex models on Apple silicon powered devices with great brevity and flexibility. We will be showcasing fine-tuning of a 7B parameter LLM on an iPhone, image generation using a large diffusion model on an iPad, and text generation using a number of large language models on a M2 Ultra Mac Studio.
- Monday, July 14: 4:00pm - 8:00pm
- Tuesday, July 15: 9:30am - 12:00pm, 2:00pm - 4:00pm
- Wednesday, July 16: 9:30am - 12:00pm, 2:00pm - 4:00pm
Accepted Papers
- "Addressing Misspecification in Simulation-based Inference through Data-driven Calibration"
- Antoine Wehenkel, Juan L. Gamella, Ozan Sener, Jens Behrmann, Guillermo Sapiro, Jörn-Henrik Jacobsen, Marco Cuturi
- "Beyond Sensor Data: Foundation Models of Behavioral Data from Wearables Improve Health Predictions"
- Eray Erturk (USC), Fahad Kamran, Salar Abbaspourazad, Sean Jewell, Harsh Sharma, Nick Foti, Joe Futoma
- "Cavia: Camera-controllable Multi-view Video Diffusion with View-Integrated Attention"
- Dejia Xu (UT Austin), Yifan Jiang, Chen (Kimi) Huang, Liangchen Song, Thorsten Gernoth, Liangliang Cao (Google), Atlas Wang (UT Austin), Hao Tang
- "CommVQ: Commutative Vector Quantization for KV Cache Compression"
- Junyan Li (University of Massachusetts Amherst), Yang Zhang (Massachusetts Institute of Technology), Muhammad Yusuf Hassan (University of Massachusetts Amherst), Talha Chafekar (University of Massachusetts Amherst), Tianle Cai (Princeton University), Zhile Ren, Pengsheng Guo, Binazir Karimzadeh, Colorado Reed, Chong Wang, Chuang Gan (University of Massachusetts Amherst)
- "Contrastive Localized Language-Image Pre-Training"
- Hong-You Chen, Jeff Lai, Haotian Zhang, Angie Wang, Marcin Eichner, Keen You, Meng Cao, Bowen Zhang, Yinfei Yang, Zhe Gan
- "CtrlSynth: Controllable Image Text Synthesis for Data-Efficient Multimodal Learning"
- Qingqing Cao, Mahyar Najibi, Sachin Mehta (Meta (work done while at Apple))
- "Distillation Scaling Laws"
- Dan Busbridge, Jason Ramapuram, Floris Weers, Amitis Shidani (University of Oxford), Russ Webb, Etai Littwin
- "Faster Rates for Private Adversarial Bandits"
- Hilal Asi, Vinod Raman (University of Michigan), Kunal Talwar
- "FlexTok: Resampling Images into 1D Token Sequences of Flexible Length"
- Jesse Allardice, Roman Bachmann, David Mizrahi, Afshin Dehghan, Elmira Amirloo, Oğuzhan Fatih Kar (EPFL), Amir Zamir (EPFL), Enrico Fini, Alaaeldin Mohamed Elnouby Ali
- "Improved Sample Complexity for Private Nonsmooth Nonconvex Optimization"
- Guy Kornowski (Weizmann Institute of Science), Daogao Liu (University of Washington), Kunal Talwar
- "INRFlow: Flow Matching for INRs in Ambient Space"
- Yuyang Wang, Anurag Ranjan, Josh Susskind, Miguel Angel Bautista Martin
- "Instruction-Following Pruning for Large Language Models"
- Bairu Hou (UCSB), Qibin Chen, Jianyu Wang, Guoli Yin, Chong Wang, Nan Du, Ruoming Pang, Shiyu Chang (UCSB), Tao Lei
- "Is Your Model Fairly Certain? A Fine-grained Approach Towards LLM Fairness"
- Oliver Wang (Carnegie Mellon University), Niv Sivakumar, Falaah Arif Khan (New York University), Rin Metcalf Susa, Adam Golinski, Natalie Mackraz, Barry Theobald, Luca Zappella, Nick Apostoloff
- "Learning to Route LLMs with Confidence Tokens"
- Allen Chuang (Rice University), Prathusha Kameswara Sarma, Parikshit Gopalan, John Boccio, Sara Bolouki, Xia (Ben) Hu (Rice University), Helen Zhou
- "Local Pan-privacy for Federated Analytics"
- Vitaly Feldman, Audra McMillan, Guy Rothblum, Kunal Talwar
- "Mechanisms of Projective Composition of Diffusion Models"
- Arwen Bradley, Preetum Nakkiran, James Thornton, David Berthelot, Josh Susskind
- "Normalizing Flows are Capable Generative Models"
- Shuangfei Zhai, Ruixiang Zhang, Preetum Nakkiran, David Berthelot, Jiatao Gu, Huangjie Zheng, Tianrong Chen, Miguel Angel Bautista Martin, Navdeep Jaitly, Josh Susskind
- "Parameters vs FLOPs: Scaling Laws for Optimal Sparsity for Mixture-of-Experts Language Models"
- Samira Abnar, Harshay Shah (MIT), Dan Busbridge, Alaaeldin Mohamed Elnouby Ali, Josh Susskind, Vimal Thilak
- "Preference Reasoning by Evaluating Decomposed preferences Inferred from Counterfactual Trajectories"
- Stéphane Aroca-Ouellette, Natalie Mackraz, Barry Theobald, Rin Metcalf Susa
- "Proxy-FDA: Proxy-based Feature Distribution Alignment for Fine-tuning Vision Foundation Models without Forgetting"
- Chen Huang, Skyler Seto, Hadi Pour Ansari, Mehrdad Farajtabar, Raviteja Vemulapalli, Fartash Faghri, Oncel Tuzel, Barry Theobald, Josh Susskind
- "QuantSpec: Self-Speculative Decoding with Hierarchical Quantized KV Cache"
- Rishabh Tiwari (UC Berkeley), Haocheng Xi (UC Berkeley), Aditya Tomar (UC Berkeley), Coleman Hooper (UC Berkeley), Sehoon Kim (UC Berkeley), Max Horton, Mahyar Najibi, Michael W. Mahoney (UC Berkeley), Kurt Keutzer (UC Berkeley), Amir Gholaminejad (UC Berkeley)
- "Robust Autonomy Emerges from Self-Play"
- Marco Cusumano-Towner, David Hafner, Alex Hertzberg, Brody Huval, Aleksei Petrenko, Eugene Vinitsky, Erik Wijmans, Taylor Killian, Stuart Bowers, Ozan Sener, Philipp Kraehenbuehl, Vladlen Koltun
- "Scaling Laws for Forgetting during Finetuning with Pretraining Data Injection"
- Louis Béthune, David Grangier, Dan Busbridge, Eleonora Gualdoni, Marco Cuturi, Pierre Ablin
- "Shielded Diffusion: Generating Novel and Diverse Images using Sparse Repellency"
- Michael Kirchhof, James Thornton, Pierre Ablin, Louis Béthune, Michal Klein, Eugene Ndiaye, Marco Cuturi
- "Soup of Experts: Pretraining Domain-Adaptive Models via Weights Averaging"
- Pierre Ablin, Angelos Katharopoulos, Skyler Seto, David Grangier
- "SPD: Sync-Point Drop for Efficient Tensor Parallelism of Large Language Models"
- Han-Byul Kim, Minsik Cho, Duc Hoang, Arnav Kundu, Mohammad Samragh Razlighi
- "Target Concrete Score Matching: A Holistic Framework for Discrete Diffusion"
- Ruixiang Zhang, Shuangfei Zhai, Yizhe Zhang, James Thornton, Zijing Ou (Imperial College London), Josh Susskind, Navdeep Jaitly
- "Tracking the Best Expert Privately"
- Hilal Asi, Aadi Saha, Vinod Raman (University of Michigan)
- "Training Software Engineering Agents and Verifiers with SWE-Gym"
- Jiayi Pan (UCB), Xingyao Wang (UIUC), Graham Neubig (CMU), Navdeep Jaitly, Ji Heng (UIUC), Alane Suhr (UCB), Yizhe Zhang
- "Understanding Input Selectivity in Mamba: Impact on Approximation Power, Memorization, and Associative Recall Capacity"
- Teresa Huang (Johns Hopkins University), Abhinav Moudgil (Concordia University, Montréal, Canada), Pau Rodriguez Lopez, Miguel Sarabia del Castillo, Luca Zappella, Federico Danieli
- "A Variational Framework for Improving Naturalness in Generative Spoken Language Models"
- Li-Wei Chen (Carnegie Mellon University), Takuya Higuchi, Zak Aldeneh, Ahmed Hussen Abdelaziz, Alexander Rudnicky (Carnegie Mellon University)
- "Variational Rectified Flow Matching"
- Pengsheng Guo, Alex Schwing
Accepted Workshop Papers
- "Boolformer: Symbolic Regression of Logic Functions with Transformers"
- Stephane d'Ascoli (EPFL), Arthur Renard (EPFL), Emmanuel Abbe, Clément Hongler (EPFL), Vassilis Joseph Papadopoulos (EPFL), Josh Susskind, Samy Bengio
- 2nd AI for Math Workshop 2025
- "Evaluating Sample Utility for Data Selection by Mimicking Model Weights"
- Brian Huang (University of Wisconsin-Madison), Manjot Bilkhu, Frederic Sala (University of Wisconsin-Madison), Javier Movellan
- DataWorld: Unifying Data Curation Frameworks Across Domains 2025
- "Lightweight Safety Guardrails via Synthetic Data and RL-guided Adversarial Training"
- Rachel Li, Riyaaz Shaik, Noyan Tokgozoglu, Aleksei Ilin, Amber Ma, Gor Matevosyan, Suhaa Dada, James Rae
- Multiple workshops (Data in Generative Models 2025, DataWorld: Unifying Data Curation Frameworks Across Domains 2025, Workshop on Reliable and Responsible Foundation Models 2025)
- "Overcoming Vocabulary Constraints with Pixel-level Fallback"
- Jonas F. Lotz (University of Copenhagen), Hendra Setiawan, Stephan Peitz, Yova Kementchedjhieva (Mohamed bin Zayed University of Artificial Intelligence)
- Tokenization Workshop 2025
- "Self-reflective Uncertainties: Do LLMs Know Their Internal Answer Distribution?"
- Michael Kirchhof, Luca Füger (independent researcher), Adam Golinski, Eeshan Gunesh Dhekane, Arno Blaas, Seong Joon Oh (University of Tübingen, Tübingen AI Center), Sinead Williamson
- Workshop on Reliable and Responsible Foundation Models 2025
- "The Geometries of Truth are Orthogonal Across Tasks"
- Waïss Azizian, Louis Béthune, Pierre Ablin, Eugene Ndiaye, Michal Klein, Michael Kirchhof, Marco Cuturi
- Workshop on Reliable and Responsible Foundation Models 2025
Acknowledgements
Natalie Schluter is a Workshop Chair for ICML.
Kunal Talwar, Marco Cuturi, Ronan Collobert, Samy Bengio, and Vladlen Koltun are Senior Area Chairs for ICML.
Byeongjoo Ahn, Chun-Liang Li, Louis Béthune, Luca Zappella, Miguel Angel Bautista Martin, Pau Rodriguez Lopez, Preetum Nakkiran, Shuangfei Zhai, Tatiana Likhomanenko, and Xavier Suau Cuadros are Area Chairs for ICML.
Dan Busbridge, Michael Kirchhof, Qingqing Cao, and Shu W. are recognized as ICML Top Reviewers.
Aakash Goel, Antoine Wehenkel, Arno Blaas, Arwen Bradley, Barry Theobald, Devon Hjelm, Eugene Ndiaye, Fahad Kamran, Fartash Faghri, Hadi Pour Ansari, Hillal Asi, Huangjie Zheng, Katherine Tsai, Manjot Bilkhu, Miguel Sarabia del Castillo, Niv Sivakumar, Pavan Kumar Anasosalu Vasu, Philipp Kraehenbuehl, Philippe Chlenski, Pierre Ablin, Rin Metcalf Susa, Tianrong Chen, Vimal Thilak, Vitaly Feldman, Xiaoming Zhao, Yinfei Yang, Yizhe Zhang, and Zhe Gan are Conference Reviewers for ICML.
Kriti Goyal is a Reviewer for the LatinX in AI Workshop.
Related readings and updates.
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2025
June 4, 2025research area Computer Vision
Apple is sponsoring the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), which will take place in person from June 11th to June 15th, 2025 in Nashville, Tennessee. CVPR is the annual computer vision event comprising the main conference and several co-located workshops and short courses. Below is the schedule of our sponsored workshops and events at CVPR 2025.
Apple will be sponsoring the International Conference on Learning Representations (ICLR), which will take place in person from April 24 to 28, 2025, in Singapore. ICLR brings together professionals dedicated to the advancement of deep learning.