Pith. sign in

REVIEW 2 major objections 5 minor 1 cited by

ParetoFlow: Guided Flows in Multi-Objective Optimization

T0 review · 2 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read ParetoFlow uses flow matching to generate designs that approximate the full Pareto front in offline multi-objective optimization.

desk verdict Solid combinatorial idea for offline MOO with flow matching, but the main SOTA claim is confounded by a predictor-based final selection step that baselines do not receive. read the letter →

arxiv 2412.03718 v2 pith:T25VUSC7 submitted 2024-12-04 cs.CE cs.AI

Ye Yuan , Can Chen , Christopher Pal , Xue Liu This is my paper · ORCID
classification cs.CEcs.AI
keywords offlinemulti-objectiveoptimizationflowmatchingParetofrontpredictorguidancegenerativemodelinglocalfilteringneighboringevolution
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

ParetoFlow asks whether a generative flow-matching model can solve offline multi-objective optimization, where the only data are pre-collected designs with multiple objective labels and no further queries to the true objective functions are allowed. The paper's claim is that it can, and that it does so better than the evolutionary, Bayesian, and other generative baselines on the Off-MOO-Bench suite. The method steers sampling by assigning each generated sample a uniform weight vector over the objectives, decomposing the multi-objective problem into many single weighted subproblems. Two additions make this work: local filtering that discards predicted objective vectors lying outside a weight-vector-aligned hypercone, and a neighboring evolution step that lets nearby weight distributions exchange offspring. If the claim holds, generative modeling becomes a practical route to approximating the entire Pareto front from offline data.

What carries the argument

The load-bearing object is the weighted multi-objective predictor guidance field. Flow matching trains a neural ODE $\hat{v}(x_t,t;\theta)$ to transport noise to data; predictor guidance adds a gradient term that pushes trajectories toward high values of a learned property. ParetoFlow replaces the single property with a weighted objective $\hat{f}_\omega(x_t;\beta)=\sum_{i=1}^m -\hat{f}_i(\hat{x}_1(x_t);\beta_i)\omega_i$, so each weight vector $\omega$ defines one guided flow. Local filtering restricts each flow to a hypercone around $\omega$, and neighboring evolution treats weight vectors within angular distance as a neighborhood whose offspring compete via the weighted-objective selection, letting similar distributions share successful intermediate states. The Pareto-optimal set update acts as a memory over the whole sampling trajectory.

What would settle it

Train the same flow model and weight schedule on a task with a known oracle, then degrade the learned predictors by adding controlled noise to their outputs during sampling; if the hypervolume of the returned 256 designs does not fall toward or below the best offline sample when predictor error rises, the paper's acknowledged dependence on accurate predictors is contradicted.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that the Pareto front can be approximated by guiding a flow-matching sampler with multi-objective predictor guidance instead of single-objective classifier guidance. Each of N samples is tied to a weight vector, and the guided vector field is $\tilde{v}(x_t,t,y;\theta)=\hat{v}(x_t,t;\theta)+\gamma \frac{1-t}{t}\nabla_{x_t}\hat{f}_\omega(x_t;\beta)$, with $\hat{f}_\omega$ the negatively weighted sum of learned objective predictors. Because uniform Das\textendash Dennis weights cover the objective simplex, the generated ensemble spans the front; the hypercone filter keeps samples on the correct Pareto segment for non-convex fronts; the neighboring update selects the best offspring among the K nearest weight distributions; and a maintained Pareto-optimal set retains the best intermediate candidates. The paper reports the best average rank across all five task groups in the benchmark and shows in ablations that each module contributes.

Load-bearing premise

The whole pipeline inherits its target from learned objective predictors, so if those predictors are wrong in the regions the sampler explores, the guided field, the filtering, and the Pareto-set updates are all misled; the paper acknowledges this in its limitation section.

Editorial extensions

If this is right

  • If the reported average ranks hold, offline multi-objective design can be treated as a generative sampling problem rather than a search problem, opening the same flow machinery used for images and molecules to engineering optimization.
  • Uniformly weighted objective distribution, rather than single-objective guidance, is the ingredient that lets one batch of samples cover the whole Pareto front.
  • Local filtering is load-bearing specifically for non-convex fronts: the ablation shows removing it barely changes the convex ZDT1 task but clearly lowers hypervolume on ZDT2.
  • Neighboring evolution turns the redundancy of similar weight vectors into a benefit, because the majority of selected offspring in the paper's ablations come from neighboring distributions rather than the sample's own distribution.
  • Retaining intermediate candidates in the Pareto-optimal set matters; the paper's ablation without the update degrades performance, so final-samples-only is not enough.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper leaves implicit that its weight-decomposition guidance is a general recipe: any conditional generative sampler that can accept a scalar guidance gradient could in principle carry the same uniform weights, local filtering, and neighbor exchange.
  • A testable extension would make the guidance uncertainty-aware, weighting or shrinking the gradient where learned predictors disagree, which directly addresses the paper's stated dependence on surrogate accuracy.
  • Adapting the hypercone angle online from local front curvature, instead of fixing it from neighbor weight distances, is a natural follow-up for strongly non-convex fronts.
  • Because the paper reports weaker generative performance on high-dimensional discrete NAS logits than on continuous tasks, the next stress test is a discrete-native flow or a better decoding scheme for architectures.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 5 minor

Summary. The manuscript proposes ParetoFlow, a flow-matching method for offline multi-objective optimization (MOO). The method decomposes the MOO task into weighted-subproblem flows: it trains per-objective predictors, guides the flow with a weighted sum of those predictors, uses a hypercone-based local filtering scheme to handle non-convex Pareto fronts, and introduces a neighboring-evolution module plus a Pareto-set memory to share information across nearby weight vectors. The authors report extensive experiments on the Off-MOO-Bench covering synthetic functions, MO-NAS, MORL, scientific design, and real-world problems, and report average-rank comparisons against evolutionary, Bayesian, and generative baselines, with ablations on the main modules.

Significance. If the empirical claims hold, ParetoFlow would be a practical and efficient generative approach to offline MOO, with a modular design that connects flow matching to decomposition-based evolutionary algorithms. The paper's strengths include broad benchmark coverage, an explicit ablation of the main modules, and a clearly stated limitation about dependence on predictor quality (Appendix A.13). However, the central state-of-the-art claim is currently clouded by a final predictor-based selection step that is not applied to baselines, so the significance of the reported advantage is not yet established by the evidence as presented.

major comments (2)
  1. [Section 4.3 and Section 3.2] The headline comparison is confounded by the final solution-selection protocol. Section 4.3 states that because Das-Dennis does not produce exactly 256 weights, the authors 'generate slightly more, resulting in over 256 samples' and then 'use learned predictors for non-dominant sorting to select the top 256 samples.' Section 3.2 repeats this in its last paragraph: 'we apply non-dominant sorting to P S and select 256 candidates for evaluation.' This is a post-hoc predictor-based truncation over an overcomplete candidate pool, and no baseline is reported to receive an analogous overgenerate-then-select wrapper; the baselines output 256 solutions directly. Because the same learned predictors guide the flow (Eq. (9)) and determine survival (Eq. (12), Algorithm 1 Line 19), the step can exploit predictor optimism, which is exactly the failure mode acknowledged in Appendix A.13. The current experiments therefore do not establish that the hypervolume advantage comes from the multi-objective guidance or neighboring evolution rather than from the final predictor-based cutoff. I request a control: apply the identical overgenerate-then-select-by-predictor protocol to an unguided flow and to a random-search baseline, or select the final 256 uniformly at random from the overgenerated pool, and report both the truncated and untruncated hypervolumes.
  2. [Section 4.5, Table 2 (w/o PS) and Section 3.2] The ablation labelled 'w/o PS' does not isolate the contribution of the Pareto-set memory or of the final selection. The text says this variant relies 'only on the final samples produced through the sampling process,' but the sample-generation process still produces more than 256 candidates (Section 4.3) and the final non-dominant sorting is performed with learned predictors (Section 3.2). The manuscript does not state whether 'w/o PS' retains the final predictor-based truncation; if it does, the ablation measures only the per-weight Pareto-set memory, not the full selection wrapper; if it does not, the number of evaluated solutions is not controlled at 256. Either way, Table 2 cannot be used to conclude that the Pareto-set update is 'critical,' because the final cutoff is a separate, uncontrolled mechanism that is common to both variants.
minor comments (5)
  1. [Appendix A.1, Eq. (16)] In Eq. (16), the right-hand side uses the symbol \tilde v(xt,t;θ) for what should be the unguided learned field \hat v(xt,t;θ); as printed, the equation is circular. Please correct the notation.
  2. [Figure 5 caption] The caption contains a typo: 'mumber of offspring O' should be 'number of offspring O.'
  3. [Section 4.4] The claim that ParetoFlow 'consistently achieves the highest ranks across all tasks' is stronger than the detailed tables support: in Tables 8 and 12, for example, DTLZ7, VLMOP1, Regex, RFP, and RE61 have other methods with higher point estimates. The claim should be phrased in terms of average rank or per-task win/loss counts.
  4. [Section 3.1, Eq. (10) and Appendix A.4] Eq. (10) adds Gaussian noise to the flow ODE, but the paper does not justify that the resulting stochastic process still samples from the intended flow-matching distribution; the noise magnitude g and the threshold rule γ=0 for t<0.8 are introduced heuristically. A brief justification or a reference to a stochastic-flow framework would improve the presentation.
  5. [Section 3.1, Local Filtering] The definition of Φ_i is ambiguous: the text says Φ_i is computed as 2 times the average of φ_ij over j, but it does not specify how φ_ij relates to the K nearest neighbors and the self-inclusion in Eq. (11). Clarifying the index ranges would help reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the guided vector field and selection rules are stated heuristics with no parameter fitted to the benchmark outcomes.

full rationale

The paper's derivation chain is self-contained and non-circular. The flow-matching model is trained with Eq. (5) on the offline dataset, and the guided vector field in Eq. (9) is a direct application of standard predictor/classifier guidance to a weighted sum of learned objective predictors; no parameter is fitted to the reported hypervolume results. The weighted distribution in Eq. (8), the local filtering scheme, and the neighboring evolution update in Eq. (12) are all stated algorithmic heuristics rather than rearrangements of the evaluation metric. The claimed SOTA in Section 4.4 is measured against a held-out ground-truth oracle, so the empirical claim is not equivalent by construction to the method's inputs. Self-citations in the paper (e.g., Chen et al. 2024, Yuan et al. 2024) appear only as related work and are not load-bearing for the central derivation. One evaluation-protocol concern is worth noting as a correctness risk rather than circularity: Section 4.3 states that the method 'generate[s] slightly more, resulting in over 256 samples' and then 'use[s] learned predictors for non-dominant sorting to select the top 256 samples,' which could give ParetoFlow an unfair advantage over baselines that do not receive this predictor-based post-filtering. However, this is an experimental confound, not a circular derivation: the final hypervolume is still computed on the oracle, and the predictors are not fitted to that oracle. The authors also explicitly acknowledge in Appendix A.13 that performance relies on predictive-model accuracy, which is a limitation but not a circularity. Thus the central claim has independent content, and no specific reduction of a prediction to its own inputs can be exhibited.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The method introduces several hand-chosen hyperparameters (gamma, g, K, O, threshold) that are not fitted to data but affect performance. The key domain assumption is the accuracy of the objective predictors. No new physical or conceptual entities are postulated; the 'weighted distribution' and 'neighboring evolution' are algorithmic constructs.

free parameters (6)
  • gamma (scaling factor) = 2 (default)
    Controls the strength of predictor guidance in Eq. (8) and Eq. (9). Set by hand; sensitivity analysis in Appendix A.3 shows performance is stable across 0-4.
  • g (noise factor) = 0.1 (default)
    Adds Gaussian noise in Eq. (10) to enhance diversity. Set by hand; sensitivity analysis in Appendix A.3.
  • K (number of neighbors) = m+1 (default)
    Number of neighboring distributions in Eq. (11). Set by hand; sensitivity analysis in Figure 4.
  • O (number of offspring) = 5 (default)
    Number of offspring per neighbor in the neighboring update. Set by hand; sensitivity analysis in Figure 5.
  • guidance threshold t = 0.8
    Time step below which gamma is set to 0 due to inaccurate reconstruction of x1_hat(xt). Determined by reconstruction loss in Figure 9.
  • Das-Dennis division parameter H = not specified
    Determines the number of uniform weight vectors; the paper states it generates 'slightly more' than 256 and then selects 256. The exact value is not given, which affects the coverage of the weight space.
assumptions (5)
  • standard math The flow matching training objective Eq. (5) learns a marginal vector field that accurately models the offline data distribution.
    The method relies on the standard flow matching framework (Lipman et al., 2023) to generate designs from the data manifold.
  • standard math Lemma 1 of Zheng et al. (2023), which justifies the predictor-guided vector field in Eq. (6), applies under the conditions used here.
    The derivation in Appendix A.1 invokes this lemma without restating its assumptions, such as the form of the conditional path and the score model.
  • domain assumption The learned objective predictors f_hat_i are sufficiently accurate to guide sampling toward the true Pareto front.
    The guidance, filtering, and Pareto set update all use these surrogates. The paper's limitation section explicitly acknowledges this reliance.
  • domain assumption The offline dataset is representative of the design space, so that flow matching can generate valid designs.
    If the dataset is sparse or biased, the generative model will not cover the relevant regions of the design space.
  • ad hoc to paper The final non-dominant sorting of the Pareto set uses only the learned predictors, not the ground-truth oracle.
    This is not explicitly stated. If the oracle is used for final selection, the offline setting is violated. This ambiguity is a red flag.

how reviews work

0 comments
Cite this review

Pith. "Pith review of ParetoFlow: Guided Flows in Multi-Objective Optimization." pith.science (2026). https://pith.science/paper/T25VUSC7

@misc{pith2026241203718,
  author       = {Pith},
  title        = {Pith review of: ParetoFlow: Guided Flows in Multi-Objective Optimization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/T25VUSC7}},
  note         = {Machine review of arXiv:2412.03718}
}
read the original abstract

In offline multi-objective optimization (MOO), we leverage an offline dataset of designs and their associated labels to simultaneously minimize multiple objectives. This setting more closely mirrors complex real-world problems compared to single-objective optimization. Recent works mainly employ evolutionary algorithms and Bayesian optimization, with limited attention given to the generative modeling capabilities inherent in such data. In this study, we explore generative modeling in offline MOO through flow matching, noted for its effectiveness and efficiency. We introduce ParetoFlow, specifically designed to guide flow sampling to approximate the Pareto front. Traditional predictor (classifier) guidance is inadequate for this purpose because it models only a single objective. In response, we propose a multi-objective predictor guidance module that assigns each sample a weight vector, representing a weighted distribution across multiple objective predictions. A local filtering scheme is introduced to address non-convex Pareto fronts. These weights uniformly cover the entire objective space, effectively directing sample generation towards the Pareto front. Since distributions with similar weights tend to generate similar samples, we introduce a neighboring evolution module to foster knowledge sharing among neighboring distributions. This module generates offspring from these distributions, and selects the most promising one for the next iteration. Our method achieves state-of-the-art performance across various tasks.

Figures

Figures reproduced from arXiv: 2412.03718 by the authors.

Figure 1
Figure 1. Motivation of Module 1 in (b) and Module 2 in (c). Distributions with similar weight vectors tend to generate similar samples. As shown in [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Local filtering: samples outside the hypercone are filtered out as shown in (c). [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Sticks and triangles are rank medians and means. 1 2 3 4 5 Number of neighbors K 0.975 0.980 0.985 0.990 0.995 1.000 1.005 1.010 1.015 Hypervolume ratio MO Hopper C 10/MOP1 [PITH_FULL_IMAGE:figures/full_fig_p017_3.png] view at source ↗
Figures from the paper (4 more)
Figure 7
Figure 7. Figure 7: Sensitivity to the scaling factor of γ. 0.025 0.05 0.1 0.2 0.4 Value of noise factor g 0.980 0.985 0.990 0.995 1.000 1.005 1.010 1.015 Hypervolume ratio MO Hopper C 10/MOP1 [PITH_FULL_IMAGE:figures/full_fig_p018_7.png]
Figure 9
Figure 9. Figure 9: Reconstruction loss as a function of the time step t. 250 500 1000 2000 4000 Number of sampling steps T 0.980 0.985 0.990 0.995 1.000 1.005 1.010 1.015 1.020 Hypervolume ratio MO Hopper C 10/MOP1 [PITH_FULL_IMAGE:figures/full_fig_p019_9.png]
Figure 11
Figure 11. Figure 11: Illustrations of ParetoFlow on two tasks C-10/MOP1 and MO-Hopper. [PITH_FULL_IMAGE:figures/full_fig_p025_11.png]
Figure 12
Figure 12. Figure 12: C-10/MOP5 case study: (1) samples prioritizing prediction error and model complexity; [PITH_FULL_IMAGE:figures/full_fig_p025_12.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. Multi-Objective-Guided Discrete Flow Matching for Controllable Biological Sequence Design

    cs.LG 2025-05 conditional novelty 6.0 of 10

    MOG-DFM uses rank-directional scoring and an adaptive hypercone filter to guide discrete flow matching toward sequences with balanced multi-objective improvements.

Reference graph

Works this paper leans on

68 extracted references · 49 canonical work pages · cited by 1 Pith paper

  1. [1]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION format.date year duplicate empty "emp...

  2. [2]

    Model-based reinforcement learning for biological sequence design

    Christof Angermueller, David Dohan, David Belanger, Ramya Deshpande, Kevin Murphy, and Lucy Colwell. Model-based reinforcement learning for biological sequence design. In International conference on learning representations, 2019

    2019

  3. [3]

    Sms-emoa: Multiobjective selection based on dominated hypervolume

    Nicola Beume, Boris Naujoks, and Michael Emmerich. Sms-emoa: Multiobjective selection based on dominated hypervolume. European Journal of Operational Research, 2007

    work page 2007

  4. [4]

    Bidirectional learning for offline infinite-width model-based optimization

    Can Chen, Yingxueff Zhang, Jie Fu, Xue Steve Liu, and Mark Coates. Bidirectional learning for offline infinite-width model-based optimization. Advances in Neural Information Processing Systems, 2022

    work page 2022

  5. [5]

    Parallel-mentoring for offline model-based optimization

    Can Chen, Christopher Beckham, Zixuan Liu, Xue Liu, and Christopher Pal. Parallel-mentoring for offline model-based optimization. In Thirty-seventh Conference on Neural Information Processing Systems, 2023 a . URL https://openreview.net/forum?id=tJwyg9Zg9G

    work page 2023

  6. [6]

    Bidirectional learning for offline model-based biological sequence design

    Can Chen, Yingxue Zhang, Xue Liu, and Mark Coates. Bidirectional learning for offline model-based biological sequence design. In International Conference on Machine Learning. PMLR, 2023 b

    work page 2023

  7. [7]

    Structure-aware protein self-supervised learning

    Can Chen, Jingbo Zhou, Fan Wang, Xue Liu, and Dejing Dou. Structure-aware protein self-supervised learning. Bioinformatics, 2023 c

    work page 2023

  8. [8]

    Robust guided diffusion for offline black-box optimization

    Can Chen, Christopher Beckham, Zixuan Liu, Xue Liu, and Christopher Pal. Robust guided diffusion for offline black-box optimization. 2024

    work page 2024

Show all 68 references
  1. [9]

    Gradnorm: Gradient normalization for adaptive loss balancing in deep multitask networks

    Zhao Chen, Vijay Badrinarayanan, Chen-Yu Lee, and Andrew Rabinovich. Gradnorm: Gradient normalization for adaptive loss balancing in deep multitask networks. In International conference on machine learning, pp.\ 794--803. PMLR, 2018

    2018

  2. [10]

    Normal-boundary intersection: A new method for generating the pareto surface in nonlinear multicriteria optimization problems

    Indraneel Das and John E Dennis. Normal-boundary intersection: A new method for generating the pareto surface in nonlinear multicriteria optimization problems. SIAM journal on optimization, 1998

    1998

  3. [11]

    Hypervolume knowledge gradient: a lookahead approach for multi-objective bayesian optimization with partial information

    Sam Daulton, Maximilian Balandat, and Eytan Bakshy. Hypervolume knowledge gradient: a lookahead approach for multi-objective bayesian optimization with partial information. In International Conference on Machine Learning. PMLR, 2023

    2023

  4. [12]

    Differentiable expected hypervolume improvement for parallel multi-objective bayesian optimization

    Samuel Daulton, Maximilian Balandat, and Eytan Bakshy. Differentiable expected hypervolume improvement for parallel multi-objective bayesian optimization. Advances in Neural Information Processing Systems, 2020

    2020

  5. [13]

    Parallel bayesian optimization of multiple noisy objectives with expected hypervolume improvement

    Samuel Daulton, Maximilian Balandat, and Eytan Bakshy. Parallel bayesian optimization of multiple noisy objectives with expected hypervolume improvement. Advances in Neural Information Processing Systems, 2021

    2021

  6. [14]

    An evolutionary many-objective optimization algorithm using reference-point-based nondominated sorting approach, part i: solving problems with box constraints

    Kalyanmoy Deb and Himanshu Jain. An evolutionary many-objective optimization algorithm using reference-point-based nondominated sorting approach, part i: solving problems with box constraints. IEEE transactions on evolutionary computation, 2013

    2013

  7. [15]

    A fast and elitist multiobjective genetic algorithm: Nsga-ii

    Kalyanmoy Deb, Amrit Pratap, Sameer Agarwal, and TAMT Meyarivan. A fast and elitist multiobjective genetic algorithm: Nsga-ii. IEEE transactions on evolutionary computation, 6, 2002

    2002

  8. [16]

    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, pp.\ 8780--8794, 2021

    2021

  9. [17]

    Nas-bench-201: Extending the scope of reproducible neural architecture search

    Xuanyi Dong and Yi Yang. Nas-bench-201: Extending the scope of reproducible neural architecture search. arXiv preprint arXiv:2001.00326, 2020

    2001 arXiv

  10. [18]

    Portfolio selection, handbook of finance, 2008

    FJ Fabozzi, HM Markowitz, and F Gupta. Portfolio selection, handbook of finance, 2008

    2008

  11. [19]

    Protgpt2 is a deep unsupervised language model for protein design

    Noelia Ferruz, Steffen Schmidt, and Birte H \"o cker. Protgpt2 is a deep unsupervised language model for protein design. Nature communications, 2022

    2022

  12. [20]

    Offline model-based optimization via normalized maximum likelihood estimation

    Justin Fu and Sergey Levine. Offline model-based optimization via normalized maximum likelihood estimation. In Proc. Int. Conf. Learning Rep. (ICLR), 2021

    2021

  13. [21]

    Protein design with guided discrete diffusion

    Nate Gruver, Samuel Stanton, Nathan Frey, Tim GJ Rudner, Isidro Hotzel, Julien Lafrance-Vanasse, Arvind Rajpal, Kyunghyun Cho, and Andrew G Wilson. Protein design with guided discrete diffusion. Advances in neural information processing systems, 2024

    2024

  14. [22]

    Training-free multi-objective diffusion model for 3d molecule generation

    Xu Han, Caihua Shan, Yifei Shen, Can Xu, Han Yang, Xiang Li, and Dongsheng Li. Training-free multi-objective diffusion model for 3d molecule generation. In The Twelfth International Conference on Learning Representations, 2023

    2023

  15. [23]

    Minimal riesz energy point configurations for rectifiable d-dimensional manifolds

    Douglas P Hardin and Edward B Saff. Minimal riesz energy point configurations for rectifiable d-dimensional manifolds. Advances in Mathematics, 2005

    2005

  16. [24]

    Joint entropy search for maximally-informed bayesian optimization

    Carl Hvarfner, Frank Hutter, and Luigi Nardi. Joint entropy search for maximally-informed bayesian optimization. Advances in Neural Information Processing Systems, 35: 0 11494--11506, 2022

    2022

  17. [25]

    Multi-objective gflownets

    Moksh Jain, Sharath Chandra Raparthy, Alex Hern \'a ndez-Garc a, Jarrid Rector-Brooks, Yoshua Bengio, Santiago Miret, and Emmanuel Bengio. Multi-objective gflownets. In International conference on machine learning, pp.\ 14631--14653. PMLR, 2023

    2023

  18. [26]

    General binding affinity guidance for diffusion models in structure-based drug design

    Yue Jian, Curtis Wu, Danny Reidenbach, and Aditi S Krishnapriyan. General binding affinity guidance for diffusion models in structure-based drug design. arXiv preprint arXiv:2406.16821, 2024

    2024

  19. [27]

    Multi-objective online learning

    Jiyan Jiang, Wenpeng Zhang, Shiji Zhou, Lihong Gu, Xiaodong Zeng, and Wenwu Zhu. Multi-objective online learning. In The Eleventh International Conference on Learning Representations, 2023. URL https://openreview.net/forum?id=dKkMnCWfVmm

    2023

  20. [28]

    Adam: A method for stochastic optimization

    Diederik P Kingma. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014

    2014 arXiv

  21. [29]

    Stochastic differential equations

    Peter E Kloeden, Eckhard Platen, Peter E Kloeden, and Eckhard Platen. Stochastic differential equations. Springer, 1992

    1992

  22. [30]

    Diffusion models as constrained samplers for optimization with unknown constraints

    Lingkai Kong, Yuanqi Du, Wenhao Mu, Kirill Neklyudov, Valentin De Bortol, Haorui Wang, Dongxia Wu, Aaron Ferber, Yi-An Ma, Carla P Gomes, et al. Diffusion models as constrained samplers for optimization with unknown constraints. arXiv preprint arXiv:2402.18012, 2024

    2024

  23. [31]

    Voicebox: Text-guided multilingual universal speech generation at scale

    Matthew Le, Apoorv Vyas, Bowen Shi, Brian Karrer, Leda Sari, Rashel Moritz, Mary Williamson, Vimal Manohar, Yossi Adi, Jay Mahadeokar, et al. Voicebox: Text-guided multilingual universal speech generation at scale. Advances in neural information processing systems, 2024

    2024

  24. [32]

    Exploring chemical space with score-based out-of-distribution generation

    Seul Lee, Jaehyeong Jo, and Sung Ju Hwang. Exploring chemical space with score-based out-of-distribution generation. In International Conference on Machine Learning. PMLR, 2023

    2023

  25. [33]

    Hw-nas-bench: Hardware-aware neural architecture search benchmark

    Chaojian Li, Zhongzhi Yu, Yonggan Fu, Yongan Zhang, Yang Zhao, Haoran You, Qixuan Yu, Yue Wang, and Yingyan Lin. Hw-nas-bench: Hardware-aware neural architecture search benchmark. arXiv preprint arXiv:2103.10584, 2021

    2021 arXiv

  26. [34]

    Evolutionary-scale prediction of atomic-level protein structure with a language model

    Zeming Lin, Halil Akin, Roshan Rao, Brian Hie, Zhongkai Zhu, Wenting Lu, Nikita Smetanin, Robert Verkuil, Ori Kabeli, Yaniv Shmueli, et al. Evolutionary-scale prediction of atomic-level protein structure with a language model. Science, 2023

    2023

  27. [35]

    Flow matching for generative modeling

    Yaron Lipman, Ricky TQ Chen, Heli Ben-Hamu, Maximilian Nickel, and Matt Le. Flow matching for generative modeling. arXiv preprint arXiv:2210.02747, 2022

    2022 arXiv

  28. [36]

    Neural architecture search as multiobjective optimization benchmarks: Problem formulation and performance assessment

    Zhichao Lu, Ran Cheng, Yaochu Jin, Kay Chen Tan, and Kalyanmoy Deb. Neural architecture search as multiobjective optimization benchmarks: Problem formulation and performance assessment. IEEE transactions on evolutionary computation, 2023

    2023

  29. [37]

    A survey of weight vector adjustment methods for decomposition-based multiobjective evolutionary algorithms

    Xiaoliang Ma, Yanan Yu, Xiaodong Li, Yutao Qi, and Zexuan Zhu. A survey of weight vector adjustment methods for decomposition-based multiobjective evolutionary algorithms. IEEE Transactions on Evolutionary Computation, 2020

    2020

  30. [38]

    Botied: Multi-objective bayesian optimization with tied multivariate ranks

    Ji Won Park, Nata s a Tagasovska, Michael Maser, Stephen Ra, and Kyunghyun Cho. Botied: Multi-objective bayesian optimization with tied multivariate ranks. arXiv preprint arXiv:2306.00344, 2023

    2023 arXiv

  31. [40]

    Multisample flow matching: Straightening flows with minibatch couplings

    Aram-Alexandre Pooladian, Heli Ben-Hamu, Carles Domingo-Enrich, Brandon Amos, Yaron Lipman, and Ricky TQ Chen. Multisample flow matching: Straightening flows with minibatch couplings. arXiv preprint arXiv:2304.14772, 2023

    2023 arXiv

  32. [41]

    Data-driven offline decision-making via invariant representation learning

    Han Qi, Yi Su, Aviral Kumar, and Sergey Levine. Data-driven offline decision-making via invariant representation learning. Advances in Neural Information Processing Systems, 2022

    2022

  33. [42]

    \ PF \ ^2 es: parallel feasible pareto frontier entropy search for multi-objective bayesian optimization

    Jixiang Qing, Henry B Moss, Tom Dhaene, and Ivo Couckuyt. \ PF \ ^2 es: parallel feasible pareto frontier entropy search for multi-objective bayesian optimization. In 26th International Conference on Artificial Intelligence and Statistcs (AISTATS) 2023, volume 206, pp.\ 2565--...

    2023

  34. [43]

    Local fitness landscape of the green fluorescent protein

    Karen S Sarkisyan, Dmitry A Bolotin, Margarita V Meer, Dinara R Usmanova, Alexander S Mishin, George V Sharonov, Dmitry N Ivankov, Nina G Bozhanova, Mikhail S Baranov, Onuralp Soylemez, et al. Local fitness landscape of the green fluorescent protein. Nature, 2016

    2016

  35. [44]

    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

    2011 arXiv

  36. [45]

    Difusco: Graph-based diffusion solvers for combinatorial optimization

    Zhiqing Sun and Yiming Yang. Difusco: Graph-based diffusion solvers for combinatorial optimization. Advances in Neural Information Processing Systems, 2023

    2023

  37. [46]

    A pareto-optimal compositional energy-based model for sampling and optimization of protein sequences

    Nata s a Tagasovska, Nathan C Frey, Andreas Loukas, Isidro H \"o tzel, Julien Lafrance-Vanasse, Ryan Lewis Kelly, Yan Wu, Arvind Rajpal, Richard Bonneau, Kyunghyun Cho, et al. A pareto-optimal compositional energy-based model for sampling and optimization of protein sequences....

    arXiv

  38. [47]

    An easy-to-use real-world multi-objective optimization problem suite

    Ryoji Tanabe and Hisao Ishibuchi. An easy-to-use real-world multi-objective optimization problem suite. Applied Soft Computing, 2020

    2020

  39. [48]

    Mujoco: A physics engine for model-based control

    Emanuel Todorov, Tom Erez, and Yuval Tassa. Mujoco: A physics engine for model-based control. In 2012 IEEE/RSJ international conference on intelligent robots and systems. IEEE, 2012

    2012

  40. [49]

    Deep Generative Modeling

    Jakub M Tomczak. Deep Generative Modeling. Springer Nature, 2022

    2022

  41. [50]

    Conservative objective models for effective offline model-based optimization

    Brandon Trabucco, Aviral Kumar, Xinyang Geng, and Sergey Levine. Conservative objective models for effective offline model-based optimization. In Proc. Int. Conf. Learning Rep. (ICLR), 2021

    2021

  42. [51]

    Design-bench: Benchmarks for data-driven offline model-based optimization

    Brandon Trabucco, Xinyang Geng, Aviral Kumar, and Sergey Levine. Design-bench: Benchmarks for data-driven offline model-based optimization. In International Conference on Machine Learning. PMLR, 2022

    2022

  43. [52]

    Multi-constraint molecular generation based on conditional transformer, knowledge distillation and reinforcement learning

    Jike Wang, Chang-Yu Hsieh, Mingyang Wang, Xiaorui Wang, Zhenxing Wu, Dejun Jiang, Benben Liao, Xujun Zhang, Bo Yang, Qiaojun He, et al. Multi-constraint molecular generation based on conditional transformer, knowledge distillation and reinforcement learning. Nature Machine Int...

    2021

  44. [53]

    Localized weighted sum method for many-objective optimization

    Rui Wang, Zhongbao Zhou, Hisao Ishibuchi, Tianjun Liao, and Tao Zhang. Localized weighted sum method for many-objective optimization. IEEE Transactions on Evolutionary Computation, 2016

    2016

  45. [54]

    Multi-objective deep data generation with correlated property control

    Shiyu Wang, Xiaojie Guo, Xuanyang Lin, Bo Pan, Yuanqi Du, Yinkai Wang, Yanfang Ye, Ashley Petersen, Austin Leitgeb, Saleh AlKhalifa, et al. Multi-objective deep data generation with correlated property control. Advances in Neural Information Processing Systems, 35: 0 28889--28...

    2022

  46. [55]

    Controllable data generation by deep learning: A review

    Shiyu Wang, Yuanqi Du, Xiaojie Guo, Bo Pan, Zhaohui Qin, and Liang Zhao. Controllable data generation by deep learning: A review. ACM Computing Surveys, 2024

    2024

  47. [56]

    Offline multi-objective optimization

    Ke Xue, Rong-Xi Tan, Xiaobin Huang, and Chao Qian. Offline multi-objective optimization. arXiv preprint arXiv:2406.03722, 2024

    2024 arXiv

  48. [57]

    Proud: Pareto-guided diffusion model for multi-objective generation

    Yinghua Yao, Yuangang Pan, Jing Li, Ivor Tsang, and Xin Yao. Proud: Pareto-guided diffusion model for multi-objective generation. Machine Learning, 2024

    2024

  49. [58]

    Roma: Robust model adaptation for offline model-based optimization

    Sihyun Yu, Sungsoo Ahn, Le Song, and Jinwoo Shin. Roma: Robust model adaptation for offline model-based optimization. In Proc. Adv. Neur. Inf. Proc. Syst (NeurIPS), 2021

    2021

  50. [59]

    Gradient surgery for multi-task learning

    Tianhe Yu, Saurabh Kumar, Abhishek Gupta, Sergey Levine, Karol Hausman, and Chelsea Finn. Gradient surgery for multi-task learning. Advances in Neural Information Processing Systems, 33: 0 5824--5836, 2020

    2020

  51. [60]

    Importance-aware co-teaching for offline model-based optimization

    Ye Yuan, Can Chen, Zixuan Liu, Willie Neiswanger, and Xue Liu. Importance-aware co-teaching for offline model-based optimization. In Proc. Adv. Neur. Inf. Proc. Syst (NeurIPS), 2023

    2023

  52. [61]

    Design editing for offline model-based optimization

    Ye Yuan, Youyuan Zhang, Can Chen, Haolun Wu, Zixuan Li, Jianmo Li, James J Clark, and Xue Liu. Design editing for offline model-based optimization. arXiv preprint arXiv:2405.13964, 2024

    2024 arXiv

  53. [62]

    Moea/d: A multiobjective evolutionary algorithm based on decomposition

    Qingfu Zhang and Hui Li. Moea/d: A multiobjective evolutionary algorithm based on decomposition. IEEE Transactions on evolutionary computation, 11, 2007

    2007

  54. [63]

    Random hypervolume scalarizations for provable multi-objective black box optimization

    Richard Zhang and Daniel Golovin. Random hypervolume scalarizations for provable multi-objective black box optimization. In International conference on machine learning. PMLR, 2020

    2020

  55. [64]

    Multi-objective optimization by learning space partitions

    Yiyang Zhao, Linnan Wang, Kevin Yang, Tianjun Zhang, Tian Guo, and Yuandong Tian. Multi-objective optimization by learning space partitions. arXiv preprint arXiv:2110.03173, 2021

    2021 arXiv

  56. [65]

    Guided flows for generative modeling and decision making

    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

    2023 arXiv

  57. [66]

    Sample-efficient multi-objective molecular optimization with GF lownets

    Yiheng Zhu, Jialu Wu, Chaowen Hu, Jiahuan Yan, Chang-Yu Hsieh, Tingjun Hou, and Jian Wu. Sample-efficient multi-objective molecular optimization with GF lownets. In Thirty-seventh Conference on Neural Information Processing Systems, 2023. URL https://openreview.net/forum?id=uoG1fLIK2s

    2023

  58. [67]

    @esa (Ref

    \@ifxundefined[1] #1\@undefined \@firstoftwo \@secondoftwo \@ifnum[1] #1 \@firstoftwo \@secondoftwo \@ifx[1] #1 \@firstoftwo \@secondoftwo [2] @ #1 \@temptokena #2 #1 @ \@temptokena \@ifclassloaded agu2001 natbib The agu2001 class already includes natbib coding, so you should ...

  59. [68]

    \@lbibitem[] @bibitem@first@sw\@secondoftwo \@lbibitem[#1]#2 \@extra@b@citeb \@ifundefined br@#2\@extra@b@citeb \@namedef br@#2 \@nameuse br@#2\@extra@b@citeb \@ifundefined b@#2\@extra@b@citeb @num @parse #2 @tmp #1 NAT@b@open@#2 NAT@b@shut@#2 \@ifnum @merge>\@ne @bibitem@firs...

  60. [69]

    Further Ablations

    @open @close @open @close and [1] URL: #1 \@ifundefined chapter * \@mkboth \@ifxundefined @sectionbib * \@mkboth * \@mkboth\@gobbletwo \@ifclassloaded amsart * \@ifclassloaded amsbook * \@ifxundefined @heading @heading NAT@ctr thebibliography [1] @ \@biblabel @NAT@ctr \@bibset...

    2024 arXiv

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.