pith. sign in

arxiv: 2512.22597 · v2 · pith:JGT7PLGKnew · submitted 2025-12-27 · 💻 cs.LG · physics.chem-ph

Energy-Guided Generative Modeling for Low-Energy Molecular Structure Discovery

Guikun Xu , Xiaohan Yi , Ziqiao Meng , Peilin Zhao , Yatao Bian This is my paper

Pith reviewed 2026-05-25 07:16 UTC · model grok-4.3

classification 💻 cs.LG physics.chem-ph
keywords molecular conformer generationenergy-guided generative modelsflow-based samplingground-state identificationlow-energy molecular structuresGEOM-QM9GEOM-Drugs
0
0 comments X

The pith

Energy-guided flow models generate accurate low-energy molecular conformers using only one or two sampling steps.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents EnFlow as a framework that integrates flow-based generation of molecular conformations with an explicit learned energy model. This coupling directs the sampling process toward low-energy regions of the conformational space, enabling both diverse ensemble generation and identification of ground-state structures. Traditional physics-based methods are computationally expensive for this task, while prior learning approaches either ignore energy calibration or produce only single structures. If the integration works as described, it yields high-fidelity conformers on benchmarks like GEOM-QM9 and GEOM-Drugs while also producing energy scores that align with physical rankings from GFN2-xTB calculations.

Core claim

EnFlow couples flow-based conformer generation with explicit energy landscape modeling to guide sampling toward low-energy regions, achieving strong performance in conformer generation and ground-state identification on GEOM-QM9 and GEOM-Drugs with only 1-2 ODE sampling steps, while the learned energy scores preserve physically meaningful energetic rankings of the generated conformations.

What carries the argument

EnFlow, the energy-guided generative framework that integrates generative flow dynamics with a learned energy model to direct sampling.

If this is right

  • Conformer ensembles can be produced with high structural fidelity under minimal ODE steps.
  • Generated conformations can be ranked by energy directly from the learned model.
  • Ground-state identification becomes possible as part of the same generative process.
  • The approach applies to both small molecules (QM9) and larger drug-like molecules (Drugs).

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The method may lower the barrier to exploring conformational landscapes in high-throughput screening settings.
  • Energy guidance could be tested for transfer to properties beyond energy, such as dipole moments or reactivity.
  • Fewer sampling steps suggest potential for scaling to larger molecular systems where full ODE integration is costly.

Load-bearing premise

That coupling generative flow dynamics with a learned energy model will reliably guide sampling to low-energy regions without introducing artifacts or requiring dataset-specific tuning beyond what is described.

What would settle it

Independent single-point quantum calculations on the generated conformations showing no correlation between the model's energy scores and actual energetic orderings, or failure to recover known low-energy structures from the benchmarks.

Figures

Figures reproduced from arXiv: 2512.22597 by Guikun Xu, Peilin Zhao, Xiaohan Yi, Yatao Bian, Ziqiao Meng.

Figure 1
Figure 1. Figure 1: (a) Potential energy landscape of molecular conformations: low-energy conformers [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: EnFlow illustrations. (a) The energy-guided flow matching framework, in which an EBM trained via the Energy Matching technique provides guidance during the flow matching process. (b) Architectural overview of the vector field and the energy model. For comparision fairness, we used the same backbone architecture as that of ET-Flow [29] (c) Illustration of improved one-step ODE sampling achieved through ener… view at source ↗
Figure 3
Figure 3. Figure 3: Joint performance on GEOM￾Drugs across molecular conformation gen￾eration and ground-state conformation pre￾diction tasks. Main Notations: A 3D molecule is for￾mally defined as M := {G, C}, where G denotes the 2D graph representation of the molecule, and C ∈ Rn×3 represents the con￾formation of the molecule in 3D space, specif￾ically encompassing the spatial coordinates of each atom. Problem Definition: Th… view at source ↗
Figure 4
Figure 4. Figure 4: Model architectures in this work. (a) Following ET-Flow [ [PITH_FULL_IMAGE:figures/full_fig_p023_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: The ablation results reveal a clear trade-off governed by the magnitude of the guidance. As the guidance strength increases, the Recall-oriented metrics (COV-R and AMR-R) exhibit a consistent degradation, whereas the Precision-oriented metrics (COV-P and AMR-P) improve substantially, with the improvements being most pronounced at small RMSD thresholds δ. At the same time, the mean predicted energy Jϕ decre… view at source ↗
Figure 5
Figure 5. Figure 5: Ablation study of guidance strengths λt for 5-step (a) and 50-step (b) ODE sampling on GEOM-QM9, and for 5-step (c) ODE sampling on GEOM-Drugs. The table reports Recall and Precision metrics at a fixed RMSD threshold of δ = 0.5 Å for GEOM-QM9 and δ = 0.75 Å for GEOM-Drugs, together with the mean predicted energy Jϕ. The plots depict how these metrics vary as a function of the RMSD threshold δ. 28 [PITH_FU… view at source ↗
Figure 6
Figure 6. Figure 6: Ablation study on the necessity of Energy Matching training. For 2-step (a) and 5-step (b) ODE sampling on GEOM-QM9, the table reports Recall and Precision metrics at a fixed RMSD threshold δ = 0.5 Å, and the plots depict how these metrics vary as a function of the RMSD threshold δ. 29 [PITH_FULL_IMAGE:figures/full_fig_p029_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Ablation study of different types of vector fields for 2-step (a) and 5-step (b) ODE [PITH_FULL_IMAGE:figures/full_fig_p030_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Ablation study of coverage (%) vs. threshold [PITH_FULL_IMAGE:figures/full_fig_p031_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Ablation study of coverage (%) vs. threshold [PITH_FULL_IMAGE:figures/full_fig_p032_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Ablation of EnsembleCert mode for ground-state conformation prediction on the GEOM-Drugs dataset. Effect of ensemble size M = 1, 5, 10, 20, 50 under 1, 2, 5, and 50 ODE sampling steps. From left to right: D-MAE (Å), D-RMSE (Å), C-RMSD (Å) [PITH_FULL_IMAGE:figures/full_fig_p033_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: With JustFM mode and 5-step ODE sampling, boxplots of ground-state confor￾mation prediction performance under three settings: (1) unguided baseline (ET-Flow; w/o guidance); (2) guided model with energy matching only (guidance & Lem); and (3) fully guided model with energy matching and energy fine-tuning (EnFlow; guidance & Lem & Lenergy). 33 [PITH_FULL_IMAGE:figures/full_fig_p033_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: (a) Six representative molecules from the GEOM-Drugs dataset. (b) Their [PITH_FULL_IMAGE:figures/full_fig_p034_12.png] view at source ↗
read the original abstract

Exploring molecular energy landscapes and identifying ground-state conformations are central challenges in computational chemistry. However, generating diverse low-energy conformers from molecular graphs remains expensive with traditional physics-based pipelines. Existing learning-based approaches remain fragmented: generative models capture conformational diversity but often lack reliable energy calibration, whereas deterministic predictors focus on a single structure and fail to represent ensemble variability. Here we introduce EnFlow, to our knowledge, the first energy-guided generative framework that couples flow-based conformer generation with explicit energy landscape modeling for joint conformational ensemble generation and ground-state identification. By integrating generative dynamics with a learned energy model, EnFlow guides sampling toward low-energy regions of the conformational landscape, improving structural fidelity under extremely few sampling steps while enabling energy-based ranking of generated conformations. Experiments on GEOM-QM9 and GEOM-Drugs show that EnFlow achieves strong performance in conformer generation and ground-state identification while requiring only 1--2 ODE sampling steps. Single-point GFN2-xTB evaluations further show that the learned energy scores preserve physically meaningful energetic rankings of generated conformations. These results support explicit energy landscape modeling as an effective strategy for low-energy molecular structure discovery through joint modeling of conformational ensembles and their associated energies.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The manuscript introduces EnFlow as the first energy-guided generative framework coupling flow-based conformer generation with explicit energy landscape modeling for joint conformational ensemble generation and ground-state identification. It reports strong performance on GEOM-QM9 and GEOM-Drugs for conformer generation and ground-state identification using only 1--2 ODE sampling steps, with learned energy scores preserving physically meaningful rankings as confirmed by single-point GFN2-xTB evaluations.

Significance. If the central claims hold, the work would be significant for computational chemistry by addressing the fragmentation between generative models (for diversity) and deterministic energy predictors through explicit joint modeling, enabling more efficient low-energy structure discovery with minimal sampling steps and direct physical validation.

major comments (1)
  1. [Abstract] Abstract: the abstract states performance claims and mentions datasets and GFN2-xTB checks but supplies no derivation details, architecture, training procedure, or quantitative metrics, preventing assessment of whether the data or integration actually supports the stated results.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their review and the opportunity to respond. We address the single major comment below.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the abstract states performance claims and mentions datasets and GFN2-xTB checks but supplies no derivation details, architecture, training procedure, or quantitative metrics, preventing assessment of whether the data or integration actually supports the stated results.

    Authors: We agree that the abstract, by design, is a concise high-level summary and therefore omits detailed derivations, architecture specifications, training procedures, and specific quantitative metrics. These elements are fully described in the Methods (Section 3) and Experiments (Section 4) sections of the manuscript. To address the concern and improve standalone readability of the abstract, we will revise it to include key quantitative metrics from the GEOM-QM9 and GEOM-Drugs results. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected from available text

full rationale

The provided abstract and context describe EnFlow as a framework coupling flow-based conformer generation with a learned energy model to guide sampling toward low-energy regions, with performance claims on GEOM datasets. No equations, derivations, or specific mechanisms (such as how energy enters the ODE or training objectives) are shown that reduce by construction to fitted inputs or self-citations. No self-definitional steps, fitted predictions, or load-bearing self-citations are identifiable. The derivation chain appears self-contained against external benchmarks like GFN2-xTB evaluations, consistent with the reader's score of 2.0 indicating no visible collapse into circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review supplies no explicit free parameters, axioms, or invented entities; the central claim rests on the unelaborated assumption that the energy model can be integrated with flow dynamics to guide sampling.

pith-pipeline@v0.9.0 · 5755 in / 1050 out tokens · 38131 ms · 2026-05-25T07:16:23.864022+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. SymDrift: One-Shot Generative Modeling under Symmetries

    cs.LG 2026-05 unverdicted novelty 6.0

    SymDrift makes drifting models produce symmetry-invariant samples in one step via symmetrized coordinate drifts or G-invariant embeddings, outperforming prior one-shot baselines on molecular benchmarks and cutting com...

Reference graph

Works this paper leans on

86 extracted references · 86 canonical work pages · cited by 1 Pith paper · 9 internal anchors

  1. [1]

    Use of 3d properties to characterize beyond rule-of-5 property space for passive permeation

    Cristiano RW Guimarães, Alan M Mathiowetz, Marina Shalaeva, Gilles Goetz, and Spiros Liras. Use of 3d properties to characterize beyond rule-of-5 property space for passive permeation. Journal of chemical information and modeling, 52(4):882–890, 2012

    work page 2012

  2. [2]

    Conformations and 3d pharmacophore searching.Drug Discovery Today: Technologies, 7(4):e245–e253, 2010

    Christof H Schwab. Conformations and 3d pharmacophore searching.Drug Discovery Today: Technologies, 7(4):e245–e253, 2010

    work page 2010

  3. [3]

    Conformation generation: the state of the art.Journal of chemical information and modeling, 57(8):1747–1756, 2017

    Paul CD Hawkins. Conformation generation: the state of the art.Journal of chemical information and modeling, 57(8):1747–1756, 2017

    work page 2017

  4. [4]

    Exploiting the potential energy landscape to sample free energy.Wiley Interdisciplinary Reviews: Computational Molecular Science, 5(3):273–289, 2015

    Andrew J Ballard, Stefano Martiniani, Jacob D Stevenson, Sandeep Somani, and David J Wales. Exploiting the potential energy landscape to sample free energy.Wiley Interdisciplinary Reviews: Computational Molecular Science, 5(3):273–289, 2015

    work page 2015

  5. [5]

    Role of molecular dynamics and related methods in drug discovery.Journal of medicinal chemistry, 59(9):4035– 4061, 2016

    Marco De Vivo, Matteo Masetti, Giovanni Bottegoni, and Andrea Cavalli. Role of molecular dynamics and related methods in drug discovery.Journal of medicinal chemistry, 59(9):4035– 4061, 2016

    work page 2016

  6. [6]

    Automated exploration of the low-energy chemical space with fast quantum chemical methods.Physical Chemistry Chemical Physics, 22(14):7169–7192, 2020

    Philipp Pracht, Fabian Bohle, and Stefan Grimme. Automated exploration of the low-energy chemical space with fast quantum chemical methods.Physical Chemistry Chemical Physics, 22(14):7169–7192, 2020

    work page 2020

  7. [7]

    Local density functional theory of atoms and molecules.Proceedings of the National Academy of Sciences, 76(6):2522–2526, 1979

    Robert G Parr, Shridhar R Gadre, and Libero J Bartolotti. Local density functional theory of atoms and molecules.Proceedings of the National Academy of Sciences, 76(6):2522–2526, 1979

    work page 1979

  8. [8]

    Glossary of terms used in physical organic chemistry.Pure Appl

    P Muller et al. Glossary of terms used in physical organic chemistry.Pure Appl. Chem, 66(5):1077–1184, 1994

    work page 1994

  9. [9]

    Springer nature, 2021

    Zhi-Hua Zhou.Machine learning. Springer nature, 2021

    work page 2021

  10. [10]

    MIT press, 2021

    Ethem Alpaydin.Machine learning. MIT press, 2021

    work page 2021

  11. [11]

    Machine learning and deep learning

    Christian Janiesch, Patrick Zschech, and Kai Heinrich. Machine learning and deep learning. Electronic markets, 31(3):685–695, 2021

    work page 2021

  12. [12]

    The autoencoding variational autoencoder.Advances in Neural Information Processing Systems, 33:15077–15087, 2020

    Taylan Cemgil, Sumedh Ghaisas, Krishnamurthy Dvijotham, Sven Gowal, and Pushmeet Kohli. The autoencoding variational autoencoder.Advances in Neural Information Processing Systems, 33:15077–15087, 2020

    work page 2020

  13. [13]

    Generative adversarial networks.Communications of the ACM, 63(11):139–144, 2020

    Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Generative adversarial networks.Communications of the ACM, 63(11):139–144, 2020

    work page 2020

  14. [14]

    Deep unsuper- vised learning using nonequilibrium thermodynamics

    Jascha Sohl-Dickstein, Eric Weiss, Niru Maheswaranathan, and Surya Ganguli. Deep unsuper- vised learning using nonequilibrium thermodynamics. InInternational conference on machine learning, pages 2256–2265. pmlr, 2015

    work page 2015

  15. [15]

    Generative modeling by estimating gradients of the data distribution.Advances in neural information processing systems, 32, 2019

    Yang Song and Stefano Ermon. Generative modeling by estimating gradients of the data distribution.Advances in neural information processing systems, 32, 2019

    work page 2019

  16. [16]

    Denoising diffusion probabilistic models.Advances in neural information processing systems, 33:6840–6851, 2020

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffusion probabilistic models.Advances in neural information processing systems, 33:6840–6851, 2020

    work page 2020

  17. [17]

    Flow matching for generative modeling

    Yaron Lipman, Ricky TQ Chen, Heli Ben-Hamu, Maximilian Nickel, and Matthew Le. Flow matching for generative modeling. InThe Eleventh International Conference on Learning Representations, 2023

    work page 2023

  18. [18]

    Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow

    Xingchao Liu, Chengyue Gong, and Qiang Liu. Flow straight and fast: Learning to generate and transfer data with rectified flow.arXiv preprint arXiv:2209.03003, 2022

    work page internal anchor Pith review Pith/arXiv arXiv 2022

  19. [19]

    Variational inference with normalizing flows

    Danilo Rezende and Shakir Mohamed. Variational inference with normalizing flows. In International conference on machine learning, pages 1530–1538. PMLR, 2015

    work page 2015

  20. [20]

    A generative model for molecular distance geometry.arXiv preprint arXiv:1909.11459, 2019

    Gregor NC Simm and José Miguel Hernández-Lobato. A generative model for molecular distance geometry.arXiv preprint arXiv:1909.11459, 2019

    work page arXiv 1909

  21. [21]

    Learning neural generative dynamics for molecular conformation generation.arXiv preprint arXiv:2102.10240, 2021

    Minkai Xu, Shitong Luo, Yoshua Bengio, Jian Peng, and Jian Tang. Learning neural generative dynamics for molecular conformation generation.arXiv preprint arXiv:2102.10240, 2021. 15

    work page arXiv 2021

  22. [22]

    An end-to-end framework for molecular conformation generation via bilevel programming

    Minkai Xu, Wujie Wang, Shitong Luo, Chence Shi, Yoshua Bengio, Rafael Gomez-Bombarelli, and Jian Tang. An end-to-end framework for molecular conformation generation via bilevel programming. InInternational conference on machine learning, pages 11537–11547. PMLR, 2021

    work page 2021

  23. [23]

    Learning gradient fields for molecular conformation generation

    Chence Shi, Shitong Luo, Minkai Xu, and Jian Tang. Learning gradient fields for molecular conformation generation. InInternational conference on machine learning, pages 9558–9568. PMLR, 2021

    work page 2021

  24. [24]

    Geomol: Torsional geometric generation of molecular 3d conformer ensembles.Advances in Neural Information Processing Systems, 34:13757–13769, 2021

    Octavian Ganea, Lagnajit Pattanaik, Connor Coley, Regina Barzilay, Klavs Jensen, William Green, and Tommi Jaakkola. Geomol: Torsional geometric generation of molecular 3d conformer ensembles.Advances in Neural Information Processing Systems, 34:13757–13769, 2021

    work page 2021

  25. [25]

    Energy-inspired molecular conformation optimization

    Jiaqi Guan, Wesley Wei Qian, Wei-Ying Ma, Jianzhu Ma, and Jian Peng. Energy-inspired molecular conformation optimization. Ininternational conference on learning representations, 2021

    work page 2021

  26. [26]

    Geodiff: A geo- metric diffusion model for molecular conformation generation.arXiv preprint arXiv:2203.02923, 2022

    Minkai Xu, Lantao Yu, Yang Song, Chence Shi, Stefano Ermon, and Jian Tang. Geodiff: A geo- metric diffusion model for molecular conformation generation.arXiv preprint arXiv:2203.02923, 2022

    work page arXiv 2022

  27. [27]

    Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point clouds

    Nathaniel Thomas, Tess Smidt, Steven Kearnes, Lusann Yang, Li Li, Kai Kohlhoff, and Patrick Riley. Tensor field networks: Rotation-and translation-equivariant neural networks for 3d point clouds.arXiv preprint arXiv:1802.08219, 2018

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  28. [28]

    E (n) equivariant graph neural networks

    Vıctor Garcia Satorras, Emiel Hoogeboom, and Max Welling. E (n) equivariant graph neural networks. InInternational conference on machine learning, pages 9323–9332. PMLR, 2021

    work page 2021

  29. [29]

    Et-flow: Equivariant flow-matching for molecular conformer generation.Advances in Neural Information Processing Systems, 37:128798–128824, 2024

    Majdi Hassan, Nikhil Shenoy, Jungyoon Lee, Hannes Stärk, Stephan Thaler, and Dominique Beaini. Et-flow: Equivariant flow-matching for molecular conformer generation.Advances in Neural Information Processing Systems, 37:128798–128824, 2024

    work page 2024

  30. [30]

    Efficient molecular conformer generation with so (3)-averaged flow matching and reflow.arXiv preprint arXiv:2507.09785, 2025

    Zhonglin Cao, Mario Geiger, Allan Dos Santos Costa, Danny Reidenbach, Karsten Kreis, Tomas Geffner, Franco Pellegrini, Guoqing Zhou, and Emine Kucukbenli. Efficient molecular conformer generation with so (3)-averaged flow matching and reflow.arXiv preprint arXiv:2507.09785, 2025

    work page arXiv 2025

  31. [31]

    Molecule3d: A benchmark for predicting 3d geometries from molecular graphs.arXiv preprint arXiv:2110.01717, 2021

    Zhao Xu, Youzhi Luo, Xuan Zhang, Xinyi Xu, Yaochen Xie, Meng Liu, Kaleb Dickerson, Cheng Deng, Maho Nakata, and Shuiwang Ji. Molecule3d: A benchmark for predicting 3d geometries from molecular graphs.arXiv preprint arXiv:2110.01717, 2021

    work page arXiv 2021

  32. [32]

    Gtmgc: Using graph transformer to predict molecule’s ground-state conformation

    Guikun Xu, Yongquan Jiang, PengChuan Lei, Yan Yang, and Jim Chen. Gtmgc: Using graph transformer to predict molecule’s ground-state conformation. InThe Twelfth International Conference on Learning Representations, 2023

    work page 2023

  33. [33]

    Bridging geometric states via geometric diffusion bridge.Advances in Neural Information Processing Systems, 37:109283–109322, 2024

    Shengjie Luo, Yixian Xu, Di He, Shuxin Zheng, Tie-Yan Liu, and Liwei Wang. Bridging geometric states via geometric diffusion bridge.Advances in Neural Information Processing Systems, 37:109283–109322, 2024

    work page 2024

  34. [34]

    Wgformer: An se (3)-transformer driven by wasserstein gradient flows for molecular ground-state conformation prediction

    Fanmeng Wang, Minjie Cheng, and Hongteng Xu. Wgformer: An se (3)-transformer driven by wasserstein gradient flows for molecular ground-state conformation prediction. InForty-second International Conference on Machine Learning, 2025

    work page 2025

  35. [35]

    Rebind: Enhancing ground- state molecular conformation prediction via force-based graph rewiring

    Taewon Kim, Hyunjin Seo, Sungsoo Ahn, and Eunho Yang. Rebind: Enhancing ground- state molecular conformation prediction via force-based graph rewiring. InThe Thirteenth International Conference on Learning Representations, 2025

    work page 2025

  36. [36]

    Do transformers really perform badly for graph representation?Advances in neural information processing systems, 34:28877–28888, 2021

    Chengxuan Ying, Tianle Cai, Shengjie Luo, Shuxin Zheng, Guolin Ke, Di He, Yanming Shen, and Tie-Yan Liu. Do transformers really perform badly for graph representation?Advances in neural information processing systems, 34:28877–28888, 2021

    work page 2021

  37. [37]

    Diffusion models beat gans on image synthesis

    Prafulla Dhariwal and Alexander Nichol. Diffusion models beat gans on image synthesis. Advances in neural information processing systems, 34:8780–8794, 2021

    work page 2021

  38. [38]

    Classifier-Free Diffusion Guidance

    Jonathan Ho and Tim Salimans. Classifier-free diffusion guidance.arXiv preprint arXiv:2207.12598, 2022. 16

    work page internal anchor Pith review Pith/arXiv arXiv 2022

  39. [39]

    Contrastive energy prediction for exact energy-guided diffusion sampling in offline reinforcement learning

    Cheng Lu, Huayu Chen, Jianfei Chen, Hang Su, Chongxuan Li, and Jun Zhu. Contrastive energy prediction for exact energy-guided diffusion sampling in offline reinforcement learning. InInternational Conference on Machine Learning, pages 22825–22855. PMLR, 2023

    work page 2023

  40. [40]

    Harmonic self-conditioned flow matching for multi-ligand docking and binding site design.arXiv preprint arXiv:2310.05764, 2023

    Hannes Stärk, Bowen Jing, Regina Barzilay, and Tommi Jaakkola. Harmonic self-conditioned flow matching for multi-ligand docking and binding site design.arXiv preprint arXiv:2310.05764, 2023

    work page arXiv 2023

  41. [41]

    A tutorial on energy-based learning.Predicting structured data, 1(0), 2006

    Yann LeCun, Sumit Chopra, Raia Hadsell, M Ranzato, Fujie Huang, et al. A tutorial on energy-based learning.Predicting structured data, 1(0), 2006

    work page 2006

  42. [42]

    Energy matching: Unifying flow matching and energy-based models for generative modeling.arXiv preprint arXiv:2504.10612, 2025

    Michal Balcerak, Tamaz Amiranashvili, Antonio Terpin, Suprosanna Shit, Lea Bogensperger, Se- bastian Kaltenbach, Petros Koumoutsakos, and Bjoern Menze. Energy matching: Unifying flow matching and energy-based models for generative modeling.arXiv preprint arXiv:2504.10612, 2025

    work page arXiv 2025

  43. [43]

    Geom, energy-annotated molecular conformations for property prediction and molecular generation.Scientific Data, 9(1):185, 2022

    Simon Axelrod and Rafael Gomez-Bombarelli. Geom, energy-annotated molecular conformations for property prediction and molecular generation.Scientific Data, 9(1):185, 2022

    work page 2022

  44. [44]

    Torsional diffusion for molecular conformer generation.Advances in neural information processing systems, 35:24240–24253, 2022

    Bowen Jing, Gabriele Corso, Jeffrey Chang, Regina Barzilay, and Tommi Jaakkola. Torsional diffusion for molecular conformer generation.Advances in neural information processing systems, 35:24240–24253, 2022

    work page 2022

  45. [45]

    Ec-conf: A ultra-fast diffusion model for molecular conformation generation with equivariant consistency.Journal of Cheminformatics, 16(1):107, 2024

    Zhiguang Fan, Yuedong Yang, Mingyuan Xu, and Hongming Chen. Ec-conf: A ultra-fast diffusion model for molecular conformation generation with equivariant consistency.Journal of Cheminformatics, 16(1):107, 2024

    work page 2024

  46. [46]

    Swallowing the bitter pill: Simplified scalable conformer generation.arXiv preprint arXiv:2311.17932, 2023

    Yuyang Wang, Ahmed A Elhag, Navdeep Jaitly, Joshua M Susskind, and Miguel Angel Bautista. Swallowing the bitter pill: Simplified scalable conformer generation.arXiv preprint arXiv:2311.17932, 2023

    work page arXiv 2023

  47. [47]

    Rdkit: A software suite for cheminformatics, computational chemistry, and predictive modeling.Greg Landrum, 8(31.10):5281, 2013

    Greg Landrum et al. Rdkit: A software suite for cheminformatics, computational chemistry, and predictive modeling.Greg Landrum, 8(31.10):5281, 2013

    work page 2013

  48. [48]

    Strategies for pre-training graph neural networks.arXiv preprint arXiv:1905.12265, 2019

    Weihua Hu, Bowen Liu, Joseph Gomes, Marinka Zitnik, Percy Liang, Vijay Pande, and Jure Leskovec. Strategies for pre-training graph neural networks.arXiv preprint arXiv:1905.12265, 2019

    work page arXiv 1905

  49. [49]

    How Attentive are Graph Attention Networks?

    Shaked Brody, Uri Alon, and Eran Yahav. How attentive are graph attention networks?arXiv preprint arXiv:2105.14491, 2021

    work page internal anchor Pith review Pith/arXiv arXiv 2021

  50. [50]

    Recipe for a general, powerful, scalable graph transformer.Advances in Neural Information Processing Systems, 35:14501–14515, 2022

    Ladislav Rampášek, Michael Galkin, Vijay Prakash Dwivedi, Anh Tuan Luu, Guy Wolf, and Dominique Beaini. Recipe for a general, powerful, scalable graph transformer.Advances in Neural Information Processing Systems, 35:14501–14515, 2022

    work page 2022

  51. [51]

    Flow Matching Guide and Code

    Yaron Lipman, Marton Havasi, Peter Holderrieth, Neta Shaul, Matt Le, Brian Karrer, Ricky TQ Chen, David Lopez-Paz, Heli Ben-Hamu, and Itai Gat. Flow matching guide and code.arXiv preprint arXiv:2412.06264, 2024

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  52. [52]

    Improving and generalizing flow-based generative models with minibatch optimal transport.Transactions on Machine Learning Research, pages 1–34, 2024

    Alexander Tong, Kilian Fatras, Nikolay Malkin, Guillaume Huguet, Yanlei Zhang, Jarrid Rector- Brooks, Guy Wolf, and Yoshua Bengio. Improving and generalizing flow-based generative models with minibatch optimal transport.Transactions on Machine Learning Research, pages 1–34, 2024

    work page 2024

  53. [53]

    On the guidance of flow matching.arXiv preprint arXiv:2502.02150, 2025

    Ruiqi Feng, Chenglei Yu, Wenhao Deng, Peiyan Hu, and Tailin Wu. On the guidance of flow matching.arXiv preprint arXiv:2502.02150, 2025

    work page arXiv 2025

  54. [54]

    Guided flows for generative modeling and decision making.arXiv preprint arXiv:2311.13443, 2023

    Qinqing Zheng, Matt Le, Neta Shaul, Yaron Lipman, Aditya Grover, and Ricky TQ Chen. Guided flows for generative modeling and decision making.arXiv preprint arXiv:2311.13443, 2023

    work page arXiv 2023

  55. [55]

    Sit: Exploring flow and diffusion-based generative models with scalable interpolant transformers

    Nanye Ma, Mark Goldstein, Michael S Albergo, Nicholas M Boffi, Eric Vanden-Eijnden, and Saining Xie. Sit: Exploring flow and diffusion-based generative models with scalable interpolant transformers. InEuropean Conference on Computer Vision, pages 23–40. Springer, 2024

    work page 2024

  56. [56]

    Diffusion Posterior Sampling for General Noisy Inverse Problems

    Hyungjin Chung, Jeongsol Kim, Michael T Mccann, Marc L Klasky, and Jong Chul Ye. Diffusion posterior sampling for general noisy inverse problems.arXiv preprint arXiv:2209.14687, 2022. 17

    work page internal anchor Pith review Pith/arXiv arXiv 2022

  57. [57]

    Loss-guided diffusion models for plug-and-play controllable generation

    Jiaming Song, Qinsheng Zhang, Hongxu Yin, Morteza Mardani, Ming-Yu Liu, Jan Kautz, Yongxin Chen, and Arash Vahdat. Loss-guided diffusion models for plug-and-play controllable generation. InInternational Conference on Machine Learning, pages 32483–32498. PMLR, 2023

    work page 2023

  58. [58]

    Eigenfold: Generative protein structure prediction with diffusion models.arXiv preprint arXiv:2304.02198, 2023

    Bowen Jing, Ezra Erives, Peter Pao-Huang, Gabriele Corso, Bonnie Berger, and Tommi Jaakkola. Eigenfold: Generative protein structure prediction with diffusion models.arXiv preprint arXiv:2304.02198, 2023

    work page arXiv 2023

  59. [59]

    Sinkhorn distances: Lightspeed computation of optimal transport.Advances in neural information processing systems, 26, 2013

    Marco Cuturi. Sinkhorn distances: Lightspeed computation of optimal transport.Advances in neural information processing systems, 26, 2013

    work page 2013

  60. [60]

    Variational analysis in the wasserstein space.arXiv preprint arXiv:2406.10676, 2024

    Nicolas Lanzetti, Antonio Terpin, and Florian Dörfler. Variational analysis in the wasserstein space.arXiv preprint arXiv:2406.10676, 2024

    work page arXiv 2024

  61. [61]

    First-order conditions for optimization in the wasserstein space.SIAM Journal on Mathematics of Data Science, 7(1):274–300, 2025

    Nicolas Lanzetti, Saverio Bolognani, and Florian Dörfler. First-order conditions for optimization in the wasserstein space.SIAM Journal on Mathematics of Data Science, 7(1):274–300, 2025

    work page 2025

  62. [62]

    The variational formulation of the fokker–planck equation.SIAM journal on mathematical analysis, 29(1):1–17, 1998

    Richard Jordan, David Kinderlehrer, and Felix Otto. The variational formulation of the fokker–planck equation.SIAM journal on mathematical analysis, 29(1):1–17, 1998

    work page 1998

  63. [63]

    Training products of experts by minimizing contrastive divergence.Neural computation, 14(8):1771–1800, 2002

    Geoffrey E Hinton. Training products of experts by minimizing contrastive divergence.Neural computation, 14(8):1771–1800, 2002

    work page 2002

  64. [64]

    Torchmd-net: equivariant transformers for neural network based molecular potentials.arXiv preprint arXiv:2202.02541, 2022

    Philipp Thölke and Gianni De Fabritiis. Torchmd-net: equivariant transformers for neural network based molecular potentials.arXiv preprint arXiv:2202.02541, 2022

    work page arXiv 2022

  65. [65]

    Uff, a full periodic table force field for molecular mechanics and molecular dynamics simulations

    Anthony K Rappé, Carla J Casewit, KS Colwell, William A Goddard III, and W Mason Skiff. Uff, a full periodic table force field for molecular mechanics and molecular dynamics simulations. Journal of the American chemical society, 114(25):10024–10035, 1992

    work page 1992

  66. [66]

    Merck molecular force field

    Thomas A Halgren. Merck molecular force field. v. extension of mmff94 using experimental data, additional computational data, and empirical rules.Journal of Computational Chemistry, 17(5-6):616–641, 1996

    work page 1996

  67. [67]

    Auto-Encoding Variational Bayes

    Diederik P Kingma and Max Welling. Auto-encoding variational bayes.arXiv preprint arXiv:1312.6114, 2013

    work page internal anchor Pith review Pith/arXiv arXiv 2013

  68. [68]

    Euclidean distance geometry and applications.SIAM review, 56(1):3–69, 2014

    Leo Liberti, Carlile Lavor, Nelson Maculan, and Antonio Mucherino. Euclidean distance geometry and applications.SIAM review, 56(1):3–69, 2014

    work page 2014

  69. [69]

    Score-Based Generative Modeling through Stochastic Differential Equations

    Yang Song, Jascha Sohl-Dickstein, Diederik P Kingma, Abhishek Kumar, Stefano Ermon, and Ben Poole. Score-based generative modeling through stochastic differential equations.arXiv preprint arXiv:2011.13456, 2020

    work page internal anchor Pith review Pith/arXiv arXiv 2011

  70. [70]

    Rdkit: Open-source cheminformatics, 2016

    rdkit. Rdkit: Open-source cheminformatics, 2016. Accessed: 2025-08-07

    work page 2016

  71. [71]

    Attention is all you need.Advances in neural information processing systems, 30, 2017

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need.Advances in neural information processing systems, 30, 2017

    work page 2017

  72. [72]

    Consistency models

    Yang Song, Prafulla Dhariwal, Mark Chen, and Ilya Sutskever. Consistency models. In International Conference on Machine Learning, pages 32211–32252. PMLR, 2023

    work page 2023

  73. [73]

    Cofm: Molecular conformation generation via flow matching in se (3)-invariant latent space

    Guikun Xu, Yankai Yu, Yongquan Jiang, Yan Yang, and Yatao Bian. Cofm: Molecular conformation generation via flow matching in se (3)-invariant latent space. InICML 2025 Generative AI and Biology (GenBio) Workshop, 2025

    work page 2025

  74. [74]

    High-resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Björn Ommer. High-resolution image synthesis with latent diffusion models. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022

    work page 2022

  75. [75]

    Ensemble kalman diffusion guidance: A derivative-free method for inverse problems.arXiv preprint arXiv:2409.20175, 2024

    Hongkai Zheng, Wenda Chu, Austin Wang, Nikola Kovachki, Ricardo Baptista, and Yisong Yue. Ensemble kalman diffusion guidance: A derivative-free method for inverse problems.arXiv preprint arXiv:2409.20175, 2024

    work page arXiv 2024

  76. [76]

    Flow matching with gaussian process priors for probabilistic time series forecasting.arXiv preprint arXiv:2410.03024, 2024

    Marcel Kollovieh, Marten Lienen, David Lüdke, Leo Schwinn, and Stephan Günnemann. Flow matching with gaussian process priors for probabilistic time series forecasting.arXiv preprint arXiv:2410.03024, 2024. 18

    work page arXiv 2024

  77. [77]

    Energy-weighted flow matching for offline reinforcement learning.arXiv preprint arXiv:2503.04975, 2025

    Shiyuan Zhang, Weitong Zhang, and Quanquan Gu. Energy-weighted flow matching for offline reinforcement learning.arXiv preprint arXiv:2503.04975, 2025

    work page arXiv 2025

  78. [78]

    Crest—a program for the exploration of low-energy molecular chemical space.The Journal of Chemical Physics, 160(11), 2024

    Philipp Pracht, Stefan Grimme, Christoph Bannwarth, Fabian Bohle, Sebastian Ehlert, Gereon Feldmann, Johannes Gorges, Marcel Müller, Tim Neudecker, Christoph Plett, et al. Crest—a program for the exploration of low-energy molecular chemical space.The Journal of Chemical Physics, 160(11), 2024

    work page 2024

  79. [79]

    Schnet–a deep learning architecture for molecules and materials.The Journal of chemical physics, 148(24), 2018

    Kristof T Schütt, Huziel E Sauceda, P-J Kindermans, Alexandre Tkatchenko, and K-R Müller. Schnet–a deep learning architecture for molecules and materials.The Journal of chemical physics, 148(24), 2018

    work page 2018

  80. [80]

    Physnet: A neural network for predicting energies, forces, dipole moments, and partial charges.Journal of chemical theory and computation, 15(6):3678–3693, 2019

    Oliver T Unke and Markus Meuwly. Physnet: A neural network for predicting energies, forces, dipole moments, and partial charges.Journal of chemical theory and computation, 15(6):3678–3693, 2019

    work page 2019

Showing first 80 references.