{"id":1172331,"date":"2026-05-26T05:34:06","date_gmt":"2026-05-26T12:34:06","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-event&p=1172331"},"modified":"2026-05-26T07:57:11","modified_gmt":"2026-05-26T14:57:11","slug":"microsoft-research-new-england-generative-modeling-sampling-seminar","status":"publish","type":"msr-event","link":"https:\/\/www.microsoft.com\/en-us\/research\/event\/microsoft-research-new-england-generative-modeling-sampling-seminar\/","title":{"rendered":"Microsoft Research New England Generative Modeling & Sampling Seminar"},"content":{"rendered":"\n\n\n\n\n

In recent years, generative models have made remarkable strides\u2014not only in generating high-fidelity images, videos, and text, but also in powering chatbots and autonomous agents. Beyond these advances, the growing connections between generative modeling and broader problems in measure transport, sampling, and stochastic control have led to exciting developments in fine-tuning, inference-time alignment, and solutions to scientific simulation and inference tasks.<\/p>\n\n\n\n

This seminar aims to bring together researchers from machine learning, statistics, applied mathematics, and the physical sciences to explore this rapidly evolving and interdisciplinary space\u2014emphasizing both theoretical foundations and practical applications.<\/p>\n\n\n\n

Nominate of self-nominate if you would like to give a talk: https:\/\/forms.gle\/xkpKqwkPerTPvW7f6 (opens in new tab)<\/span><\/a> <\/p>\n\n\n\n

Based in Microsoft Research New England<\/a> (MSR NE), we welcome a diverse set of speakers and participants and aim to foster cross-disciplinary conversations and collaborations. The seminar takes place in the New England \/ Boston area with in-person participation from the local research community and virtual attendance from a global audience. We host speakers in person whenever possible. <\/p>\n\n\n\n

This seminar is currently organized by Carles Domingo-Enrich (opens in new tab)<\/span><\/a> and Yuanqi Du (opens in new tab)<\/span><\/a> at Microsoft Research New England, and first established by Carles Domingo-Enrich (opens in new tab)<\/span><\/a>, Dinghuai Zhang (opens in new tab)<\/span><\/a>, and Yuanqi Du (opens in new tab)<\/span><\/a>. <\/p>\n\n\n\n

Online attendance<\/h2>\n\n\n\n

Fill out this 1-minute form (opens in new tab)<\/span><\/a> (if it doesn’t work try it on a different browser\/device). The email you input will be added to the Google Group for the seminar, which means you will get Teams invites and emails about upcoming talks. It may take a few hours (up to a day) for you to get the welcome email.<\/p>\n\n\n\n

In-person attendance<\/h2>\n\n\n\n

If you are a Microsoft New England employee, just show up. If you are an external guest, fill out this 1-minute form (opens in new tab)<\/span><\/a> (if it doesn’t work try it on a different browser\/device). Provide your name and ID to the building reception at the entrance, take the elevators to the M floor, and ask the receptionist at the M floor for a visitor tag and for Room Clara Barton. Bring a personal ID!<\/strong><\/p>\n\n\n\n

If you have any questions, send an email to carlesd@microsoft.com<\/a> and yuanqidu@microsoft.com<\/a>.<\/p>\n\n\n\n

\n
Teams link<\/a><\/div>\n<\/div>\n\n\n\n

Schedule<\/h2>\n\n\n\n
Date<\/th>Time<\/th>Speaker in-person?<\/th>Speaker<\/th>Title<\/th><\/tr><\/thead>
June 23, 2026<\/td>10:30 AM<\/td>Yes<\/td>Yunyi Shen, MIT<\/td><\/td><\/tr>
June 16, 2026<\/td>10:30 AM<\/td>Yes<\/td>Yuanqi Du, Microsoft Research
Carles Domingo-Enrich, Microsoft Research<\/td>		Rare Event Analysis via Stochastic Optimal Control<\/td><\/tr>
June 9, 2026<\/td>10:30 AM<\/td>Yes<\/td>Michelle M. Li, Harvard University<\/td><\/td><\/tr>
June 2, 2026<\/td>10:30 AM<\/td>Yes<\/td>Mouyang Cheng, MIT<\/td><\/td><\/tr>
May 26, 2026<\/td>10:30 AM<\/td>Yes<\/td>Runzhong Wang, MIT<\/td>De novo Generation for Molecular Structure Elucidation from Mass Spectrometry<\/td><\/tr>
May 19, 2026<\/td>10:30 AM<\/td>Yes<\/td>Aimee Maurais, MIT<\/td>Designing Dynamic Measure Transport for Sampling | Watch<\/a><\/td><\/tr>
May 12, 2026<\/td>10:30 AM<\/td>Yes<\/td>Lingkai Kong, Harvard University<\/td>Generative AI for High-Stakes Decision-Making with Applications in One Health | Watch<\/a><\/td><\/tr>
May 5, 2026<\/td>10:30 AM<\/td>No<\/td>Luca Ambrogioni\u200b, Donders Institute for Brain, Cognition and Behaviour<\/td>How Out-of-Equilibrium Phase Transitions can Seed Pattern Formation in Trained Diffusion Models | Watch<\/a><\/td><\/tr>
April 28, 2026<\/td>10:30 AM<\/td>Yes<\/td>Emma Finn, Harvard University<\/td>Where the Score Lives: What Wavelets Reveal About Diffusion Models | Watch<\/a><\/td><\/tr>
April 21, 2026<\/td><\/td><\/td>No talk (ICLR week)<\/td><\/td><\/tr>
April 14, 2026<\/td>10:30 AM<\/td>Yes<\/td>Mujin Kwun, Harvard University
Carles Domingo-Enrich, Microsoft Research<\/td>		Matching Features, Not Tokens: Energy-Based Fine-Tuning of Language Models | Watch<\/a><\/td><\/tr>
April 9, 2026<\/td>4:00 PM<\/td>Yes<\/td>Yixuan Wang, Caltech<\/td>Data-Driven Discovery and Verification of Singularities in Nonlinear Partial Differential Equations | Watch<\/a><\/td><\/tr>
April 7, 2026<\/td>11:00 AM<\/td>No<\/td>Tristan Bereau, Heidelberg University<\/td>Tractable Mapping Entropy and Generative Backmapping via Split-Flows | Watch<\/a><\/td><\/tr>
March 31, 2026<\/td>10:30 AM<\/td>No<\/td>Michael Plainer, ELIZA Zuse School
Winfried Ripken, TU Berlin
Gregor Lied, TU Berlin<\/td>		Generative Models for Molecular Dynamics Across Timescales | Watch<\/a><\/td><\/tr>
March 24, 2026<\/td>11:00 AM<\/td>No<\/td>Yinuo Ren, Stanford University<\/td>A Unified Approach to Analysis and Design of Denoising Markov Models | Watch<\/a><\/td><\/tr>
March 17, 2026<\/td>11:00 AM<\/td>No<\/td>Qiyang (Colin) Li, UC Berkeley<\/td>Q-learning with Flow-Matching Policies | Watch<\/a><\/td><\/tr>
March 10, 2026<\/td>10:30 AM<\/td>No<\/td>Jiequn Han, Flatiron Institute<\/td>Generative Modeling without Clean Data: Self-Consistent Transport under Black-Box Corruptions | Watch<\/a><\/td><\/tr>
March 3, 2026<\/td>10:30 AM<\/td>Yes<\/td>David Layden, IBM Research<\/td>Wavefunction Flows: Efficient Quantum Simulation of Continuous Flow Models | Watch<\/a><\/td><\/tr>
February 24, 2026<\/td>10:30 AM<\/td>Yes<\/td>Lorenz Richter, dida, Zuse Institute Berlin<\/td>A non-Markovian approach to diffusion-based sampling | Watch<\/a><\/td><\/tr>
February 17, 2026<\/td>10:30 AM<\/td>Yes<\/td>Aram-Alexandre Pooladian, Yale University<\/td>Blind denoising diffusion models and the blessings of dimensionality | Watch<\/a><\/td><\/tr>
February 10, 2026<\/td><\/td><\/td>No talk (break)<\/td><\/td><\/tr>
February 3, 2026<\/td>10:30 AM<\/td>Yes<\/td>Peter Potaptchik, University of Oxford, Harvard University<\/td>Meta Flow Maps | Watch<\/a><\/td><\/tr>
January 27, 2026<\/td><\/td><\/td>No talk (ICML deadline)<\/td><\/td><\/tr>
January 20, 2026<\/td>10:30 AM<\/td>No<\/td>Luca Eyring, Technical University of Munich<\/td>Reward fine-tuning of Few-step Diffusion Models with Noise Hypernetworks<\/td><\/tr>
January 13, 2026<\/td>10:30 AM<\/td>Yes<\/td>Aayush Karan, Harvard University<\/td>Reasoning with Sampling: Your Base Model is Smarter Than You Think<\/td><\/tr>
January 6, 2026<\/td>11:00 AM<\/td>No<\/td>Fan Chen, MIT<\/td>The Coverage Principle: How Pre-Training Enables Post-Training<\/td><\/tr>
December 30, 2025<\/td><\/td><\/td>No talk (winter break)<\/td><\/td><\/tr>
December 23, 2025<\/td><\/td><\/td>No talk (winter break)<\/td><\/td><\/tr>
December 16, 2025<\/td><\/td><\/td>No talk (winter break)<\/td><\/td><\/tr>
December 9, 2025<\/td>12:00 PM<\/td>Yes<\/td>Runqian Wang, MIT<\/td>Equilibrium Matching: Generative Modeling with Implicit Energy-Based Models<\/td><\/tr>
December 2, 2025<\/td><\/td><\/td>No talk (NeurIPS)<\/td><\/td><\/tr>
November 25, 2025<\/td>10:00 AM<\/td>Yes<\/td>Peter Holderrieth, MIT<\/td>GLASS Flows: Transition Sampling for Alignment of Flow and Diffusion Models<\/td><\/tr>
November 18, 2025<\/td>11:00 AM<\/td>No<\/td>Jorge L. Rosa-Raices, UC Berkeley<\/td>Nonadiabatic flow matching for free-energy estimation in and out of equilibrium<\/td><\/tr>
November 11, 2025<\/td>11:00 AM<\/td>Yes<\/td>Peter Potaptchik, University of Oxford, Harvard University<\/td>Tilt Matching for Scalable Sampling and Fine-Tuning<\/td><\/tr>
November 4, 2025<\/td>11:00 AM<\/td>Yes<\/td>Alex Berlaga, University of Chicago<\/td>Targeting Low-Energy Protein Ensembles with Adjoint Matching<\/td><\/tr>
October 28, 2025<\/td>11:00 AM<\/td>No<\/td>Denis Blessing, Karlsruhe<\/td>Sampling with trust region constraints<\/td><\/tr>
October 21, 2025<\/td>11:00 AM<\/td>Yes<\/td>Chin-Wei Huang, Microsoft Research<\/td>Accurate and scalable density functional with deep learning<\/td><\/tr>
October 14, 2025<\/td>11:00 AM<\/td>Yes<\/td>Jaeyeon Kim, Harvard University
Brian Lee Cheuk-Kit, Harvard University<\/td>		Any-Order, Any-Length, Any-Time: Extending Masked Diffusion Models with Flexibility and Self-Correction<\/td><\/tr>
October 7, 2025<\/td>11:00 AM<\/td>No<\/td>Adam Block, Columbia University<\/td>EMA Without the Lag: Stabilizing Optimization for Behavior Cloning and Language Models<\/td><\/tr>
September 30, 2025<\/td>11:00 AM<\/td>Yes<\/td>Ava Amini, Kevin K. Yang, Microsoft Research<\/td>The Dayhoff Atlas: scaling sequence diversity for improved protein generation<\/td><\/tr>
September 23, 2025<\/td><\/td><\/td>No talk (ICLR deadline)<\/td><\/td><\/tr>
September 16, 2025<\/td>11:00 AM<\/td>Yes<\/td>Songlin Yang, MIT<\/td>Toward More Expressive yet Scalable RNNs: DeltaNet and Its Variants<\/td><\/tr>
September 9, 2025<\/td>11:00 AM<\/td>Yes<\/td>Zongyi Li, MIT<\/td>Neural Operator for Scientific Computing<\/td><\/tr>
September 2, 2025<\/td>11:00 AM<\/td>Yes<\/td>Kwangjun Ahn, Microsoft Research<\/td>Dion: The distributed orthonormal update revolution is here<\/td><\/tr>
August 26, 2025<\/td>11:00 AM<\/td>Yes<\/td>Zhengyang Geng, Carnegie Mellon University<\/td>Mean Flows for One-Step Generative Modeling<\/td><\/tr>
August 19, 2025<\/td>11:00 AM<\/td>No<\/td>Jiajun He, University of Cambridge
Yuanqi Du, Cornell, Microsoft Research
Francisco Vargas, University of Cambridge, Xaira<\/td>		Fantastic Path RND and Where to Find Them<\/td><\/tr>
August 12, 2025<\/td>11:00 AM<\/td>Yes<\/td>Luhuan Wu, Flatiron<\/td>A training-free diffusion-based SMC sampler for unnormalized distributions<\/td><\/tr>
August 5, 2025<\/td>11:00 AM<\/td>Yes<\/td>Raghav Singhal, NYU, Microsoft Research<\/td>Inference-Time Steering of diffusion models<\/td><\/tr>
July 29, 2025<\/td>11:00 AM<\/td>Yes<\/td>Sitan Chen, Harvard University<\/td>What does guidance do?<\/td><\/tr>
July 22, 2025<\/td>11:00 AM<\/td>No<\/td>Guan-Horng Liu, Meta FAIR<\/td>Sampling with Schr\u00f6dinger Bridge \u2014 An Adjoint-Matching Perspective<\/td><\/tr>
July 15, 2025<\/td>11:00 AM<\/td>Yes<\/td>Tianhong Li, MIT<\/td>Broadening the Scope of Autoregressive Models in Computer Vision and Beyond<\/td><\/tr>
July 8, 2025<\/td>11:00 AM<\/td>No<\/td>Jiaxin Shi, Google Deepmind<\/td>Discrete Generative Modeling with Masked Diffusions<\/td><\/tr>
July 1, 2025<\/td>11:00 AM<\/td>Yes<\/td>Yilun Du, Harvard University<\/td>Inference Time Reasoning with Generative Models<\/td><\/tr>
June 24, 2025<\/td>11:00 AM<\/td>Yes<\/td>Michael Albergo, Harvard University<\/td>Generative Flow Maps via Self-distillation<\/td><\/tr>
June 17, 2025<\/td>11:00 AM<\/td>No<\/td>Kirill Neklyudov, MILA<\/td>Solving Many-Body Schr\u00f6dinger Equation from the Probabilistic Perspective<\/td><\/tr>
June 10, 2025<\/td>11:00 AM<\/td>Yes<\/td>Marta Skreta, University of Toronto, Microsoft Research<\/td>Controlling Diffusion Models at Inference Time<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n","protected":false},"excerpt":{"rendered":"

In recent years, generative models have made remarkable strides\u2014not only in generating high-fidelity images, videos, and text, but also in powering chatbots and autonomous agents. Beyond these advances, the growing connections between generative modeling and broader problems in measure transport, sampling, and stochastic control have led to exciting developments in fine-tuning, inference-time alignment, and solutions […]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr_startdate":"2025-06-10","msr_enddate":"2026-09-29","msr_location":"One Memorial Drive, Cambridge, MA, Room Clara Barton on the M floor","msr_expirationdate":"","msr_event_recording_link":"","msr_event_link":"","msr_event_link_redirect":false,"msr_event_time":"10:30 AM - 11:30 PM every Tuesday","msr_hide_region":false,"msr_private_event":false,"msr_hide_image_in_river":0,"footnotes":""},"research-area":[13556],"msr-region":[256048,197900],"msr-event-type":[197944],"msr-video-type":[],"msr-locale":[268875],"msr-program-audience":[],"msr-post-option":[],"msr-impact-theme":[],"class_list":["post-1172331","msr-event","type-msr-event","status-publish","hentry","msr-research-area-artificial-intelligence","msr-region-global","msr-region-north-america","msr-event-type-hosted-by-microsoft","msr-locale-en_us"],"msr_about":"\n\n\n\n\n

In recent years, generative models have made remarkable strides\u2014not only in generating high-fidelity images, videos, and text, but also in powering chatbots and autonomous agents. Beyond these advances, the growing connections between generative modeling and broader problems in measure transport, sampling, and stochastic control have led to exciting developments in fine-tuning, inference-time alignment, and solutions to scientific simulation and inference tasks.<\/p>\n\n\n\n

This seminar aims to bring together researchers from machine learning, statistics, applied mathematics, and the physical sciences to explore this rapidly evolving and interdisciplinary space\u2014emphasizing both theoretical foundations and practical applications.<\/p>\n\n\n\n

Nominate of self-nominate if you would like to give a talk: https:\/\/forms.gle\/xkpKqwkPerTPvW7f6<\/a> <\/p>\n\n\n\n

Based in Microsoft Research New England<\/a> (MSR NE), we welcome a diverse set of speakers and participants and aim to foster cross-disciplinary conversations and collaborations. The seminar takes place in the New England \/ Boston area with in-person participation from the local research community and virtual attendance from a global audience. We host speakers in person whenever possible. <\/p>\n\n\n\n

This seminar is currently organized by Carles Domingo-Enrich<\/a> and Yuanqi Du<\/a> at Microsoft Research New England, and first established by Carles Domingo-Enrich<\/a>, Dinghuai Zhang<\/a>, and Yuanqi Du<\/a>. <\/p>\n\n\n\n

Online attendance<\/h2>\n\n\n\n

Fill out this 1-minute form<\/a> (if it doesn't work try it on a different browser\/device). The email you input will be added to the Google Group for the seminar, which means you will get Teams invites and emails about upcoming talks. It may take a few hours (up to a day) for you to get the welcome email.<\/p>\n\n\n\n

In-person attendance<\/h2>\n\n\n\n

If you are a Microsoft New England employee, just show up. If you are an external guest, fill out this 1-minute form<\/a> (if it doesn't work try it on a different browser\/device). Provide your name and ID to the building reception at the entrance, take the elevators to the M floor, and ask the receptionist at the M floor for a visitor tag and for Room Clara Barton. Bring a personal ID!<\/strong><\/p>\n\n\n\n

If you have any questions, send an email to carlesd@microsoft.com<\/a> and yuanqidu@microsoft.com<\/a>.<\/p>\n\n\n\n

\n
Teams link<\/a><\/div>\n<\/div>\n\n\n\n

Schedule<\/h2>\n\n\n\n
Date<\/th>Time<\/th>Speaker in-person?<\/th>Speaker<\/th>Title<\/th><\/tr><\/thead>
June 23, 2026<\/td>10:30 AM<\/td>Yes<\/td>Yunyi Shen, MIT<\/td><\/td><\/tr>
June 16, 2026<\/td>10:30 AM<\/td>Yes<\/td>Yuanqi Du, Microsoft Research
Carles Domingo-Enrich, Microsoft Research<\/td>		Rare Event Analysis via Stochastic Optimal Control<\/td><\/tr>
June 9, 2026<\/td>10:30 AM<\/td>Yes<\/td>Michelle M. Li, Harvard University<\/td><\/td><\/tr>
June 2, 2026<\/td>10:30 AM<\/td>Yes<\/td>Mouyang Cheng, MIT<\/td><\/td><\/tr>
May 26, 2026<\/td>10:30 AM<\/td>Yes<\/td>Runzhong Wang, MIT<\/td>De novo Generation for Molecular Structure Elucidation from Mass Spectrometry<\/td><\/tr>
May 19, 2026<\/td>10:30 AM<\/td>Yes<\/td>Aimee Maurais, MIT<\/td>Designing Dynamic Measure Transport for Sampling | Watch<\/a><\/td><\/tr>
May 12, 2026<\/td>10:30 AM<\/td>Yes<\/td>Lingkai Kong, Harvard University<\/td>Generative AI for High-Stakes Decision-Making with Applications in One Health | Watch<\/a><\/td><\/tr>
May 5, 2026<\/td>10:30 AM<\/td>No<\/td>Luca Ambrogioni\u200b, Donders Institute for Brain, Cognition and Behaviour<\/td>How Out-of-Equilibrium Phase Transitions can Seed Pattern Formation in Trained Diffusion Models | Watch<\/a><\/td><\/tr>
April 28, 2026<\/td>10:30 AM<\/td>Yes<\/td>Emma Finn, Harvard University<\/td>Where the Score Lives: What Wavelets Reveal About Diffusion Models | Watch<\/a><\/td><\/tr>
April 21, 2026<\/td><\/td><\/td>No talk (ICLR week)<\/td><\/td><\/tr>
April 14, 2026<\/td>10:30 AM<\/td>Yes<\/td>Mujin Kwun, Harvard University
Carles Domingo-Enrich, Microsoft Research<\/td>		Matching Features, Not Tokens: Energy-Based Fine-Tuning of Language Models | Watch<\/a><\/td><\/tr>
April 9, 2026<\/td>4:00 PM<\/td>Yes<\/td>Yixuan Wang, Caltech<\/td>Data-Driven Discovery and Verification of Singularities in Nonlinear Partial Differential Equations | Watch<\/a><\/td><\/tr>
April 7, 2026<\/td>11:00 AM<\/td>No<\/td>Tristan Bereau, Heidelberg University<\/td>Tractable Mapping Entropy and Generative Backmapping via Split-Flows | Watch<\/a><\/td><\/tr>
March 31, 2026<\/td>10:30 AM<\/td>No<\/td>Michael Plainer, ELIZA Zuse School
Winfried Ripken, TU Berlin
Gregor Lied, TU Berlin<\/td>		Generative Models for Molecular Dynamics Across Timescales | Watch<\/a><\/td><\/tr>
March 24, 2026<\/td>11:00 AM<\/td>No<\/td>Yinuo Ren, Stanford University<\/td>A Unified Approach to Analysis and Design of Denoising Markov Models | Watch<\/a><\/td><\/tr>
March 17, 2026<\/td>11:00 AM<\/td>No<\/td>Qiyang (Colin) Li, UC Berkeley<\/td>Q-learning with Flow-Matching Policies | Watch<\/a><\/td><\/tr>
March 10, 2026<\/td>10:30 AM<\/td>No<\/td>Jiequn Han, Flatiron Institute<\/td>Generative Modeling without Clean Data: Self-Consistent Transport under Black-Box Corruptions | Watch<\/a><\/td><\/tr>
March 3, 2026<\/td>10:30 AM<\/td>Yes<\/td>David Layden, IBM Research<\/td>Wavefunction Flows: Efficient Quantum Simulation of Continuous Flow Models | Watch<\/a><\/td><\/tr>
February 24, 2026<\/td>10:30 AM<\/td>Yes<\/td>Lorenz Richter, dida, Zuse Institute Berlin<\/td>A non-Markovian approach to diffusion-based sampling | Watch<\/a><\/td><\/tr>
February 17, 2026<\/td>10:30 AM<\/td>Yes<\/td>Aram-Alexandre Pooladian, Yale University<\/td>Blind denoising diffusion models and the blessings of dimensionality | Watch<\/a><\/td><\/tr>
February 10, 2026<\/td><\/td><\/td>No talk (break)<\/td><\/td><\/tr>
February 3, 2026<\/td>10:30 AM<\/td>Yes<\/td>Peter Potaptchik, University of Oxford, Harvard University<\/td>Meta Flow Maps | Watch<\/a><\/td><\/tr>
January 27, 2026<\/td><\/td><\/td>No talk (ICML deadline)<\/td><\/td><\/tr>
January 20, 2026<\/td>10:30 AM<\/td>No<\/td>Luca Eyring, Technical University of Munich<\/td>Reward fine-tuning of Few-step Diffusion Models with Noise Hypernetworks<\/td><\/tr>
January 13, 2026<\/td>10:30 AM<\/td>Yes<\/td>Aayush Karan, Harvard University<\/td>Reasoning with Sampling: Your Base Model is Smarter Than You Think<\/td><\/tr>
January 6, 2026<\/td>11:00 AM<\/td>No<\/td>Fan Chen, MIT<\/td>The Coverage Principle: How Pre-Training Enables Post-Training<\/td><\/tr>
December 30, 2025<\/td><\/td><\/td>No talk (winter break)<\/td><\/td><\/tr>
December 23, 2025<\/td><\/td><\/td>No talk (winter break)<\/td><\/td><\/tr>
December 16, 2025<\/td><\/td><\/td>No talk (winter break)<\/td><\/td><\/tr>
December 9, 2025<\/td>12:00 PM<\/td>Yes<\/td>Runqian Wang, MIT<\/td>Equilibrium Matching: Generative Modeling with Implicit Energy-Based Models<\/td><\/tr>
December 2, 2025<\/td><\/td><\/td>No talk (NeurIPS)<\/td><\/td><\/tr>
November 25, 2025<\/td>10:00 AM<\/td>Yes<\/td>Peter Holderrieth, MIT<\/td>GLASS Flows: Transition Sampling for Alignment of Flow and Diffusion Models<\/td><\/tr>
November 18, 2025<\/td>11:00 AM<\/td>No<\/td>Jorge L. Rosa-Raices, UC Berkeley<\/td>Nonadiabatic flow matching for free-energy estimation in and out of equilibrium<\/td><\/tr>
November 11, 2025<\/td>11:00 AM<\/td>Yes<\/td>Peter Potaptchik, University of Oxford, Harvard University<\/td>Tilt Matching for Scalable Sampling and Fine-Tuning<\/td><\/tr>
November 4, 2025<\/td>11:00 AM<\/td>Yes<\/td>Alex Berlaga, University of Chicago<\/td>Targeting Low-Energy Protein Ensembles with Adjoint Matching<\/td><\/tr>
October 28, 2025<\/td>11:00 AM<\/td>No<\/td>Denis Blessing, Karlsruhe<\/td>Sampling with trust region constraints<\/td><\/tr>
October 21, 2025<\/td>11:00 AM<\/td>Yes<\/td>Chin-Wei Huang, Microsoft Research<\/td>Accurate and scalable density functional with deep learning<\/td><\/tr>
October 14, 2025<\/td>11:00 AM<\/td>Yes<\/td>Jaeyeon Kim, Harvard University
Brian Lee Cheuk-Kit, Harvard University<\/td>		Any-Order, Any-Length, Any-Time: Extending Masked Diffusion Models with Flexibility and Self-Correction<\/td><\/tr>
October 7, 2025<\/td>11:00 AM<\/td>No<\/td>Adam Block, Columbia University<\/td>EMA Without the Lag: Stabilizing Optimization for Behavior Cloning and Language Models<\/td><\/tr>
September 30, 2025<\/td>11:00 AM<\/td>Yes<\/td>Ava Amini, Kevin K. Yang, Microsoft Research<\/td>The Dayhoff Atlas: scaling sequence diversity for improved protein generation<\/td><\/tr>
September 23, 2025<\/td><\/td><\/td>No talk (ICLR deadline)<\/td><\/td><\/tr>
September 16, 2025<\/td>11:00 AM<\/td>Yes<\/td>Songlin Yang, MIT<\/td>Toward More Expressive yet Scalable RNNs: DeltaNet and Its Variants<\/td><\/tr>
September 9, 2025<\/td>11:00 AM<\/td>Yes<\/td>Zongyi Li, MIT<\/td>Neural Operator for Scientific Computing<\/td><\/tr>
September 2, 2025<\/td>11:00 AM<\/td>Yes<\/td>Kwangjun Ahn, Microsoft Research<\/td>Dion: The distributed orthonormal update revolution is here<\/td><\/tr>
August 26, 2025<\/td>11:00 AM<\/td>Yes<\/td>Zhengyang Geng, Carnegie Mellon University<\/td>Mean Flows for One-Step Generative Modeling<\/td><\/tr>
August 19, 2025<\/td>11:00 AM<\/td>No<\/td>Jiajun He, University of Cambridge
Yuanqi Du, Cornell, Microsoft Research
Francisco Vargas, University of Cambridge, Xaira<\/td>		Fantastic Path RND and Where to Find Them<\/td><\/tr>
August 12, 2025<\/td>11:00 AM<\/td>Yes<\/td>Luhuan Wu, Flatiron<\/td>A training-free diffusion-based SMC sampler for unnormalized distributions<\/td><\/tr>
August 5, 2025<\/td>11:00 AM<\/td>Yes<\/td>Raghav Singhal, NYU, Microsoft Research<\/td>Inference-Time Steering of diffusion models<\/td><\/tr>
July 29, 2025<\/td>11:00 AM<\/td>Yes<\/td>Sitan Chen, Harvard University<\/td>What does guidance do?<\/td><\/tr>
July 22, 2025<\/td>11:00 AM<\/td>No<\/td>Guan-Horng Liu, Meta FAIR<\/td>Sampling with Schr\u00f6dinger Bridge \u2014 An Adjoint-Matching Perspective<\/td><\/tr>
July 15, 2025<\/td>11:00 AM<\/td>Yes<\/td>Tianhong Li, MIT<\/td>Broadening the Scope of Autoregressive Models in Computer Vision and Beyond<\/td><\/tr>
July 8, 2025<\/td>11:00 AM<\/td>No<\/td>Jiaxin Shi, Google Deepmind<\/td>Discrete Generative Modeling with Masked Diffusions<\/td><\/tr>
July 1, 2025<\/td>11:00 AM<\/td>Yes<\/td>Yilun Du, Harvard University<\/td>Inference Time Reasoning with Generative Models<\/td><\/tr>
June 24, 2025<\/td>11:00 AM<\/td>Yes<\/td>Michael Albergo, Harvard University<\/td>Generative Flow Maps via Self-distillation<\/td><\/tr>
June 17, 2025<\/td>11:00 AM<\/td>No<\/td>Kirill Neklyudov, MILA<\/td>Solving Many-Body Schr\u00f6dinger Equation from the Probabilistic Perspective<\/td><\/tr>
June 10, 2025<\/td>11:00 AM<\/td>Yes<\/td>Marta Skreta, University of Toronto, Microsoft Research<\/td>Controlling Diffusion Models at Inference Time<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n","tab-content":[],"msr_startdate":"2025-06-10","msr_enddate":"2026-09-29","msr_event_time":"10:30 AM - 11:30 PM every Tuesday","msr_location":"One Memorial Drive, Cambridge, MA, Room Clara Barton on the M floor","msr_event_link":"","msr_event_recording_link":"","msr_startdate_formatted":"June 10, 2025","msr_register_text":"Watch now","msr_cta_link":"","msr_cta_text":"","msr_cta_bi_name":"","featured_image_thumbnail":null,"event_excerpt":"In recent years, generative models have made remarkable strides\u2014not only in generating high-fidelity images, videos, and text, but also in powering chatbots and autonomous agents. Beyond these advances, the growing connections between generative modeling and broader problems in measure transport, sampling, and stochastic control have led to exciting developments in fine-tuning, inference-time alignment, and solutions to scientific simulation and inference tasks. This seminar aims to bring together researchers from machine learning, statistics, applied mathematics, and…","msr_research_lab":[199563],"related-researchers":[{"type":"user_nicename","display_name":"Carles Domingo-Enrich","user_id":43632,"people_section":"Section name 0","alias":"carlesd"}],"msr_impact_theme":[],"related-academic-programs":[],"related-groups":[],"related-projects":[],"related-opportunities":[],"related-publications":[],"related-videos":[1172594,1172597,1172600,1172603,1172606,1172609,1172612,1172615,1172618,1172621,1172624,1172627,1172630,1172633,1173310],"related-posts":[],"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event\/1172331","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-event"}],"version-history":[{"count":13,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event\/1172331\/revisions"}],"predecessor-version":[{"id":1173475,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event\/1172331\/revisions\/1173475"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=1172331"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=1172331"},{"taxonomy":"msr-region","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-region?post=1172331"},{"taxonomy":"msr-event-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event-type?post=1172331"},{"taxonomy":"msr-video-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video-type?post=1172331"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=1172331"},{"taxonomy":"msr-program-audience","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-program-audience?post=1172331"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=1172331"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=1172331"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}