Papers and talks


Home / Papers and talks
Esmeralda S. Whitammer Papers and presentations
Publications and preprints Work that first appeared before 2026 uses my old name (with a couple of exceptions in the immediate vicinity of the change). All citations should reflect the name on the most recent version of paper, as shown in the list below. See this page for details. Please also see my Google Scholar, Semantic Scholar, and ORCID profiles. Unpublished preprints (last 6 months) None at the moment. 2026 Accepted / published A comedy of estimators: On KL regularization in RL training of LLMs Vedant Shah, Johan Obando-Ceron, Vineet Jain, Brian Bartoldson, Bhavya Kailkhura, Sarthak Mittal, Glen Berseth, Pablo Samuel Castro, Yoshua Bengio, Esmeralda S. Whitammer, Moksh Jain, Siddarth Venkatraman, Aaron Courville COLM 2026 Forgetting is everywhere Ben Sanati, Thomas Lee, Trevor McInroe, Aidan Scannell, Esmeralda S. Whitammer, David Abel, Amos Storkey CoLLAs 2026 Lightweight latent reasoning for narrative tasks Alex Gurung, Esmeralda S. Whitammer, Mirella Lapata TACL Likelihood hacking in probabilistic program synthesis Jacek Karwowski, Younesse Kaddar, Zihuiwen Ye, Esmeralda S. Whitammer, Sam Staton UAI 2026 Bayesian symbolic regression with entropic reinforcement learning Oussama Boussif, Mohammed Mahfoud, Younesse Kaddar, Moksh Jain, Sida Li, Damiano Fornasiere, Xiaoyin Chen, Yoshua Bengio, Esmeralda S. Whitammer UAI 2026 Imperfect world models are exploitable Logan Mondal Bhamidipaty, Esmeralda S. Whitammer, David Abel, Mykel J. Kochenderfer, Subramanian Ramamoorthy RLC 2026 “Reinforcement Learning in Big Worlds” workshop Path-dependent discrete amortised inference Tiago Silva, Esmeralda S. Whitammer, Salem Lahlou ICML 2026 Discrete diffusion samplers and bridges: Off-policy algorithms and applications in latent spaces Arran Carter, Sanghyeok Choi, Kirill Tamogashev, Víctor Elvira, Esmeralda S. Whitammer ICML 2026 Reinforced sequential Monte Carlo for amortised sampling Sanghyeok Choi, Sarthak Mittal, Víctor Elvira, Jinkyoo Park, Esmeralda S. Whitammer ICML 2026 Generalised latent slice sampling Kirill Tamogashev, Sanghyeok Choi, Arran Carter, Víctor Elvira, Alice Doucet Beaupré, Esmeralda S. Whitammer ICML 2026 “Structured Probabilistic Inference and Generative Modeling” workshop Stop the sampler! Classifier-based adaptive stopping for sampling kernels Kirill Korolev, Nikita Morozov, Stepan Pavlenko, Esmeralda S. Whitammer, Sergey Samsonov ICML 2026 “Structured Probabilistic Inference and Generative Modeling” workshop Structured inference with large language Gibbs Sanghyeok Choi, Henry Gouk, Esmeralda S. Whitammer ICML 2026 “Structured Probabilistic Inference and Generative Modeling” workshop Amortised inference through one-step implicit sampling Vincent Pauline, Kirill Tamogashev, Arran Carter, Sanghyeok Choi, Stefan Bauer, Esmeralda S. Whitammer ICML 2026 “Structured Probabilistic Inference and Generative Modeling” workshop How to approximate inference with subtractive mixture models Lena Zellinger, Nicola Branchini, Lennert De Smet, Víctor Elvira, N.M., Antonio Vergari AISTATS 2026 Multi-marginal flow matching with adversarially learnt interpolants Oskar Kviman, Kirill Tamogashev, Nicola Branchini, Víctor Elvira, Jens Lagergren, N.M. ICLR 2026 Data-to-energy stochastic dynamics Kirill Tamogashev, N.M. ICLR 2026 Discrete compositional generation via general soft operators and robust reinforcement learning Marco Jiralerspong, Esther Derman, Danilo Vucetic, N.M., Bilun Sun, Tianyu Zhang, Pierre-Luc Bacon, Gauthier Gidel ICLR 2026 Structured image representation learning for flow-matching models Alexandros Graikos, Kostas Triaridis, N.M., Dimitris Samaras ICLR 2026 “Deep Generative Model in Machine Learning” workshop From discrete-time policies to continuous-time diffusion samplers: Asymptotic equivalences and faster training Julius Berner, Lorenz Richter, Marcin Sendera, Jarrid Rector-Brooks, N.M. TMLR 2025 Accepted / published Recursive self-aggregation unlocks deep thinking in large language models Siddarth Venkatraman, Vineet Jain, Sarthak Mittal, Vedant Shah, Johan Obando-Ceron, Yoshua Bengio, Brian Bartoldson, Bhavya Kailkhura, Guillaume Lajoie, Glen Berseth, N.M., Moksh Jain preprint NeurIPS 2025 “Foundations of Reasoning in Language Models” workshop Adaptive destruction processes for diffusion samplers Timofei Gritsaev, Nikita Morozov, Kirill Tamogashev, Daniil Tiapkin, Sergey Samsonov, Alexey Naumov, Dmitry Vetrov, N.M. preprint NeurIPS 2025 “Frontiers in Probabilistic Inference” workshop Fast flow-based visuomotor policies via conditional optimal transport couplings Andreas Sochopoulos, N.M., Nikolaos Tsagkas, João Moura, Michael Gienger, Sethu Vijayakumar CoRL 2025 Mixtures of in-context learners Giwon Hong, Emile Van Krieken, N.M., Edoardo Ponti, Pasquale Minervini ACL 2025 Can a Bayesian oracle prevent harm from an agent? Yoshua Bengio, Michael K. Cohen, N.M., Matt MacDermott, Damiano Fornasiere, Pietro Greiner, Younesse Kaddar UAI 2025 Outsourced diffusion sampling: Efficient posterior inference in latent spaces of generative models Siddarth Venkatraman, Mohsin Hasan, Minsu Kim, Luca Scimeca, Marcin Sendera, Yoshua Bengio, Glen Berseth, N.M. ICML 2025 On designing diffusion autoencoders for efficient generation and representation learning Magdalena Proszewska, N.M., N. Siddharth preprint CVPR 2025 “Generative Models for Computer Vision” workshop Action abstractions for amortized sampling Oussama Boussif, Léna Néhale Ezzine, Joseph Viviano, Michał Koziarski, Moksh Jain, N.M., Emmanuel Bengio, Rim Assouel, Yoshua Bengio ICLR 2025 Adaptive teachers for amortized samplers Minsu Kim, Sanghyeok Choi, Taeyoung Yun, Emmanuel Bengio, Leo Feng, Jarrid Rector-Brooks, Sungsoo Ahn, Jinkyoo Park, N.M., Yoshua Bengio ICLR 2025 Learning diverse attacks on large language models for robust red-teaming and safety tuning Seanie Lee, Minsu Kim, Lynn Cherif, David Dobre, Juho Lee, Sung Ju Hwang, Kenji Kawaguchi, Gauthier Gidel, Yoshua Bengio, N.M., Moksh Jain ICLR 2025 PQMass: Probabilistic assessment of the quality of generative models using probability mass estimation Pablo Lemos, Sammy Nasser Sharief, N.M., Laurence Perreault-Levasseur, Yashar Hezaveh ICLR 2025 Learning decision trees as amortized structure inference Mohammed Mahfoud, Ghait Boukachab, Michał Koziarski, Alex Hernández-García, Stefan Bauer, Yoshua Bengio, N.M. ICLR 2025 “Frontiers in Probabilistic Inference” workshop Preprints / notes Pixellated posterior sampling of point spread functions in astronomical images Connor Stone, Ronan Legin, Alexandre Adam, N.M., Gabriel Missael Barco, Laurence Perreault-Levasseur, Yashar Hezaveh preprint Mind the information gap: Unveiling detailed morphologies of z∼0.5–1.0 galaxies with SLACS strong lenses and data-driven analysis Ronan Legin, Connor Stone, Alexandre Adam, Gabriel Missael Barco, Adam Coogan, N.M., Laurence Perreault-Levasseur, Yashar Hezaveh preprint In-context parametric inference: Point or distribution estimators? Sarthak Mittal, Yoshua Bengio, N.M., Guillaume Lajoie preprint 2024 Accepted / published Amortizing intractable inference in diffusion models for vision, language, and control Siddarth Venkatraman, Moksh Jain, Luca Scimeca, Minsu Kim, Marcin Sendera, Mohsin Hasan, Luke Rowe, Sarthak Mittal, Pablo Lemos, Emmanuel Bengio, Alexandre Adam, Jarrid Rector-Brooks, Yoshua Bengio, Glen Berseth, N.M. NeurIPS 2024 Improved off-policy training of diffusion samplers Marcin Sendera, Minsu Kim, Sarthak Mittal, Pablo Lemos, Luca Scimeca, Jarrid Rector-Brooks, Alexandre Adam, Yoshua Bengio, N.M. NeurIPS 2024 Proof Flow: Preliminary study on generative flow network language model tuning for formal reasoning Matthew Ho, Vincent Zhu, Xiaoyin Chen, Moksh Jain, N.M., Edwin Zhang NeurIPS 2024 “System-2 Reasoning at Scale” workshop Amortizing intractable inference in diffusion models for Bayesian inverse problems [extension of conference paper] Siddarth Venkatraman, Moksh Jain, Luca Scimeca, Minsu Kim, Marcin Sendera, Mohsin Hasan, Luke Rowe, Sarthak Mittal, Pablo Lemos, Emmanuel Bengio, Alexandre Adam, Jarrid Rector-Brooks, Yashar Hezaveh, Laurence Perreault-Levasseur, Yoshua Bengio, Glen Berseth, N.M. NeurIPS 2024 “Machine Learning and the Physical Sciences” workshop Path-filtering in path-integral simulations of open quantum systems using GFlowNets Jeremy Lackman-Mincoff, Moksh Jain, N.M., Yoshua Bengio, Lena Simine Journal of Chemical Physics 161(14), 2024 V-STaR: Training verifiers for self-taught reasoners Arian Hosseini, Xingdi Yuan, N.M., Aaron Courville, Alessandro Sordoni, Rishabh Agarwal COLM 2024 Machine learning and information theory concepts towards an AI Mathematician Yoshua Bengio, N.M. Bulletin of the American Mathematical Society, 2024 Iterated denoising energy matching for sampling from Boltzmann densities Tara Akhound-Sadegh, Jarrid Rector-Brooks, Joey Bose, Sarthak Mittal, Pablo Lemos, Cheng-Hao Liu, Marcin Sendera, Siamak Ravanbakhsh, Gauthier Gidel, Yoshua Bengio, N.M., Alexander Tong ICML 2024 Improving gradient-guided nested sampling for posterior inference Pablo Lemos, N.M., Will Handley, Yoshua Bengio, Yashar Hezaveh, Laurence Perreault-Levasseur ICML 2024 Discrete probabilistic inference as control in multi-path environments Tristan Deleu, Padideh Nouri, N.M., Doina Precup, Yoshua Bengio UAI 2024 Amortizing intractable inference in large language models Edward Hu, Moksh Jain, Eric Elmoznino, Younesse Kaddar, Guillaume Lajoie, Yoshua Bengio, N.M. ICLR 2024; best paper honourable mention Delta-AI: Local objectives for amortized inference in sparse graphical models Jean-Pierre Falet, Hae-Beom Lee, N.M., Chen Sun, Dragos Secrieru, Dinghuai Zhang, Guillaume Lajoie, Yoshua Bengio ICLR 2024 Expected flow networks in stochastic environments and two-player zero-sum games Marco Jiralerspong, Bilun Sun, Danilo Vucetic, Tianyu Zhang, Yoshua Bengio, Gauthier Gidel, N.M. ICLR 2024 PhyloGFN: Phylogenetic inference with GFlowNets Ming Yang Zhou, Zichao Yan, Elliot Layne, N.M., Dinghuai Zhang, Moksh Jain, Mathieu Blanchette, Yoshua Bengio ICLR 2024 Simulation-free Schrödinger bridges via score and flow matching Alexander Tong, N.M., Kilian Fatras, Lazar Atanackovic, Yanlei Zhang, Guillaume Huguet, Hananeh Aliee, Guy Wolf, Yoshua Bengio AISTATS 2024 Improving and generalizing flow-based generative models with minibatch optimal transport Alexander Tong, Kilian Fatras, N.M., Guillaume Huguet, Yanlei Zhang, Jarrid Rector-Brooks, Guy Wolf, Yoshua Bengio TMLR Preprints / notes On generalization for generative flow networks Anas Krichel, N.M., Salem Lahlou, Yoshua Bengio preprint 2023 Accepted / published Joint Bayesian inference of graphical structure and parameters with a single generative flow network Tristan Deleu, Mizu Nishikawa-Toomey, Jithendaraa Subramanian, N.M., Laurent Charlin, Yoshua Bengio NeurIPS 2023 Let the flows tell: Solving graph combinatorial problems with GFlowNets Dinghuai Zhang, Hanjun Dai, N.M., Aaron Courville, Yoshua Bengio, Ling Pan NeurIPS 2023 Donor activity is associated with US legislators’ attention to political issues Pranav Goel, N.M., SoRelle Gaynor, Nebojsa Jojic, Kristina Miler, Philip Resnik PLOS One, 2023 GFlowNet-EM for learning compositional latent variable models Edward Hu, N.M., Moksh Jain, Katie Everett, Alexandros Graikos, Yoshua Bengio ICML 2023 A theory of continuous generative flow networks Salem Lahlou, Tristan Deleu, Pablo Lemos, Dinghuai Zhang, Alexandra Volokhova, Alex Hernández-García, Léna Néhale Ezzine, Yoshua Bengio, N.M. ICML 2023 Learning GFlowNets from partial episodes for improved convergence and stability Kanika Madan, Jarrid Rector-Brooks, Maksym Korablyov, Emmanuel Bengio, Moksh Jain, Andrei Nica, Tom Bosc, Yoshua Bengio, N.M. ICML 2023 Better training of GFlowNets with local credit and incomplete trajectories Ling Pan, N.M., Dinghuai Zhang, Yoshua Bengio ICML 2023 GFlowOut: Dropout with generative flow networks Dianbo Liu, Moksh Jain, Bonaventure Dossou, Qianli Shen, Salem Lahlou, Anirudh Goyal, N.M., Chris Emezue, Dinghuai Zhang, Nadhir Hassen, Xu Ji, Kenji Kawaguchi, Yoshua Bengio ICML 2023 Thompson sampling for improved exploration in GFlowNets Jarrid Rector-Brooks, Kanika Madan, Moksh Jain, Maksym Korablyov, Cheng-Hao Liu, Sarath Chandar, N.M., Yoshua Bengio ICML 2023 “Structured Probabilistic Inference and Generative Modeling” workshop BatchGFN: Generative flow networks for batch active learning Shreshth Malik, Salem Lahlou, Andrew Jesson, Moksh Jain, N.M., Tristan Deleu, Yoshua Bengio, Yarin Gal ICML 2023 “Structured Probabilistic Inference and Generative Modeling” workshop Probabilistic reasoning over sets using large language models Batu Ozturkler, N.M., Zhen Wang, Nebojsa Jojic ACL 2023 GFlowNets and variational inference N.M., Salem Lahlou, Tristan Deleu, Xu Ji, Edward Hu, Katie Everett, Dinghuai Zhang, Yoshua Bengio ICLR 2023 Preprints / notes Unifying generative models with GFlowNets and beyond Dinghuai Zhang, Ricky T. Q. Chen, N.M., Yoshua Bengio preprint 2022 Accepted / published Diffusion models as plug-ang-play priors Alexandros Graikos, N.M., Nebojsa Jojic, Dimitris Samaras NeurIPS 2022 Trajectory balance: Improved credit assignment in GFlowNets N.M., Moksh Jain, Emmanuel Bengio, Chen Sun, Yoshua Bengio NeurIPS 2022 Posterior samples of source galaxies in strong gravitational lenses with score-based priors Alexandre Adam, Adam Coogan, N.M., Ronan Legin, Laurence Perreault Levasseur, Yashar Hezaveh, Yoshua Bengio NeurIPS 2022 “Machine Learning for the Physical Sciences” workshop Resolving label uncertainty with implicit posterior models Esther Rolf, N.M.*, Alexandros Graikos, Ana Jojic, Caleb Robinson, Nebojsa Jojic UAI 2022 Generative flow networks for discrete probabilistic modeling Dinghuai Zhang, N.M., Zhen Liu, Alexandra Volokhova, Aaron Courville, Yoshua Bengio ICML 2022 Unifying generative models with GFlowNets Dinghuai Zhang, Ricky T. Q. Chen, N.M., Yoshua Bengio ICML 2022 “Beyond Bayes: Paths Towards Universal Reasoning Systems” workshop Coherence boosting: When your pretrained language model is not paying enough attention N.M., Zhen Wang, Nebojsa Jojic ACL 2022 The outcome of the 2021 IEEE GRSS Data Fusion Contest - Track MSD: Multitemporal semantic change detection Zhuohong Li, Fangxiao Lu, Hongyan Zhang, Lilin Tu, Jiayi Li, Xin Huang, Caleb Robinson, N.M., Nebojsa Jojic, Pedram Ghamisi, Ronny Hänsch, Naoto Yokoya JSTARS vol.15 2021 Accepted / published Studying word order through iterative shuffling N.M., Sameera Lanka, Pranav Goel, Nebojsa Jojic EMNLP 2021 GPT Perdetry Test: Generating new meanings for new words N.M., Sameera Lanka, Pranav Goel, Sudha Rao, Nebojsa Jojic NAACL 2021 From local algorithms to global results: Human-machine collaboration for robust analysis of geographically diverse imagery Nebojsa Jojic, N.M., Caleb Robinson, Anthony Ortiz IGARSS 2021 On the Galois action on motivic fundamental groups of punctured elliptic and rational curves N.M.; thesis advisor A.B. Goncharov PhD thesis Global land cover mapping with weak supervision: Outcome of the 2020 IEEE GRSS Data Fusion Contest Caleb Robinson, N.M., Nebojsa Jojic, Huijun Chen, Rongjun Qin, Changlin Xiao, Michael Schmitt, Pedram Ghamisi, Ronny Hänsch, Naoto Yokoya JSTARS vol.14 Preprints / notes High-resolution land cover change from low-resolution labels: Simple baselines for the 2021 IEEE GRSS Data Fusion Contest N.M., Caleb Robinson, Nebojsa Jojic preprint 2020 Accepted / published Weakly supervised semantic segmentation in the 2020 IEEE GRSS Data Fusion Contest Caleb Robinson, N.M., Lucas Hu, Bistra Dilkina, Nebojsa Jojic IGARSS 2020; contest winner Mining self-similarity: Label super-resolution with epitomic representations N.M., Anthony Ortiz, Nebojsa Jojic ECCV 2020 Human-machine collaboration for fast land cover mapping Caleb Robinson, Anthony Ortiz, N.M., Blake Elias, Andi Peng, Dan Morris, Bistra Dilkina, Nebojsa Jojic AAAI 2020 Preprints / notes Learning intersecting representations of short random walks on graphs N.M., Nebojsa Jojic preprint Motivic fundamental groups of CM elliptic curves and geometry of Bianchi hyperbolic threefolds N.M. preprint Shuffle relations for Hodge and motivic correlators N.M. preprint 2019 Accepted / published Large scale high-resolution land cover mapping with multi-resolution data Caleb Robinson, Le Hou, N.M., Rachel Soobitsky, Jacob Czawlytko, Bistra Dilkina, Nebojsa Jojic CVPR 2019 Label super-resolution networks N.M., Caleb Robinson, Le Hou, Rachel Soobitsky, Jacob Czawlytko, Dimitris Samaras, Joel Saltz, Lucas Joppa, Nebojsa Jojic ICLR 2019 Preprints / notes Label super-resolution with inter-instance loss Maozheng Zhao, Le Hou, Han Le, Dimitris Samaras, Nebojsa Jojic, Danielle Fassler, Tahsin Kurc, Rajarsi Gupta, N.M., Shroyer Kenneth, Joel Saltz preprint		Invited talks I intend to eventually post slides for some of these. 2026 Language models as priors and proposals for probabilistic inference Jagiellonian University, Machine Learning Summer School on Reliability and Safety Kraków, July 2026 Dynamics-based generative modelling (mini-course) SciCADE Research School Edinburgh, June 2026 Dynamic samplers and bridges with amortised sequential decision-making University of Manchester, Centre for AI Fundamentals Manchester, May 2026 How to sample (and more) with dynamic measure transport University of Edinburgh, School of Mathematics, Numerical Mathematics and Data Science seminar Edinburgh, January 2026 2025 How to sample (and more) with dynamic measure transport Constructor University, Bayes Group Bremen, November 2025 Learning to construct for generative modelling and Bayesian inference Constructor University Bremen, November 2025 How to sample (and more) with dynamic measure transport Technische Universität München Munich, November 2025 Deep dynamic measure transport: from amortised sampling to Schrödinger bridges and beyond Internationales Wissenschaftsforum, Generative models in science and machine learning workshop Heidelberg, September 2025 Diffusion modelling for amortised inference Isaac Newton Institute for Mathematical Sciences, Accelerating statistical inference and experimental design with machine learning workshop Cambridge, June 2025 Diffusion model tutorial Isaac Newton Institute for Mathematical Sciences, Accelerating statistical inference and experimental design with machine learning workshop Cambridge, June 2025 Stochastic control for black-box inference: Insights from deep reinforcement learning BayesComp 2025 Singapore, June 2025 Amortised inference meets LLMs: Algorithms and implications for faithful knowledge extraction Simons Institute for the Theory of Computing, Safety-Guaranteed LLMs workshop Berkeley, April 2025 The latest in generative AI research Edinburgh Futures Institute / Morgan Stanley Inclusive Ventures Lab Edinburgh, February 2025 Diffusion models without data: Towards plans and representations from stochastic dynamics ETH Zürich, AI Center Zürich, February 2025 2024 Plans and symbolic representations from stochastic dynamics CIFAR, Learning in Machines and Brains program meeting Toronto, November 2024 Diffusion models without data KTH Royal Institute of Technology, Digital Futures Stockholm, October 2024 Amortising intractable inference with diffusion models and off-policy RL Higher School of Economics, High-Dimensional Inference Lab Moscow / virtual, August 2024 2023 Developments in amortized posterior inference with foundation models CIFAR, Learning in Machines and Brains program meeting New Orleans, December 2023 Two variational perspectives on diffusion models Google DeepMind London / virtual, December 2023 Bayesian neurosymbolic AI for reasoning and scientific discovery University of Edinburgh, School of Informatics Edinburgh, November 2023 Generative flow networks for inference over structured objects Stony Brook University, Computer Science colloquium Stony Brook, March 2023 Probabilistic inference for reasoning with large language models Columbia University, NLP seminar New York, March 2023 Probabilistic inference for reasoning with large language models Microsoft Research Montréal / virtual, February 2023 Generative flow networks: Theory, applications, and connections Google Research, Bayesflow seminar New York / virtual, January 2023 2022 Coherence boosting: When your pretrained language model is not paying enough attention ACL 2022 Dublin, May 2022 2021 Studying word order through iterative shuffling EMNLP 2021 Punta Cana, November 2021 GPT Perdetry Test: Generating new meanings for new words NAACL 2021 virtual, June 2021 New approaches to computer vision for land cover mapping and change detection Microsoft Research Redmond / virtual, March 2021 2020 Motivic fundamental groups of CM elliptic curves and geometry of Bianchi hyperbolic threefolds Johns Hopkins University, Junior Number Theory Days Baltimore / virtual, December 2020 Land cover mapping with epitomes and clustering models ML for Remote Sensing seminar virtual, August 2020 Human-machine collaboration for fast land cover mapping ICLR 2020, Climate Change AI virtual, April 2020