Improving Inverse Folding for Peptide Design with Diversity-regularized Direct Preference Optimization

Bruno Trentini; C. Brian Roland; Chen Tessler; Darren J. Hsu; Maria Korshunova; Olivia Viessmann; Ryan Park; Shie Mannor

arxiv: 2410.19471 · v1 · submitted 2024-10-25 · 💻 cs.LG · cs.AI

Improving Inverse Folding for Peptide Design with Diversity-regularized Direct Preference Optimization

Ryan Park , Darren J. Hsu , C. Brian Roland , Maria Korshunova , Chen Tessler , Shie Mannor , Olivia Viessmann , Bruno Trentini This is my paper

Pith reviewed 2026-05-23 19:01 UTC · model grok-4.3

classification 💻 cs.LG cs.AI

keywords inverse foldingpeptide designdirect preference optimizationProteinMPNNdiversity regularizationstructural similaritysequence diversitymachine learning

0 comments

The pith

Diversity-regularized DPO fine-tunes ProteinMPNN to generate more varied peptide sequences that match target structures better than the base model.

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

The paper establishes that inverse folding models like ProteinMPNN tend to output repetitive sequences for peptides that fail to fold into given reference structures. It shows that direct preference optimization can be enhanced with online diversity regularization and domain-specific priors to produce sequences that are both more diverse and more structurally consistent. A sympathetic reader would care because this directly addresses a practical bottleneck in structure-based peptide design, where both variety for screening and fidelity to a desired fold matter. When conditioned on OpenFold-generated structures, the approach yields at least an 8% gain in structural similarity over the base model and up to 20% higher sequence diversity than plain DPO with no drop in similarity.

Core claim

When conditioned on OpenFold generated structures, fine-tuned models achieve state-of-the-art structural similarity scores, improving base ProteinMPNN by at least 8%. Compared to standard DPO, the regularized method achieves up to 20% higher sequence diversity with no loss in structural similarity score.

What carries the argument

Diversity-regularized Direct Preference Optimization, which adds online diversity regularization and domain-specific priors to standard DPO training on decoder models.

If this is right

Fine-tuned models achieve at least 8% higher structural similarity scores than base ProteinMPNN on OpenFold-conditioned peptide tasks.
The regularized method delivers up to 20% higher sequence diversity than standard DPO while preserving structural similarity.
The enhancements apply specifically to decoder-based inverse folding for short peptide sequences.
Online diversity regularization improves variety in generated sequences without requiring changes to the underlying model architecture.

Where Pith is reading between the lines

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

The same regularization could be tested on longer protein chains where repetition is also observed in inverse folding outputs.
Combining the method with experimental validation loops might accelerate identification of functional peptides from computational candidates.
The reliance on OpenFold suggests the pipeline could be closed by feeding predicted structures back into design iterations.

Load-bearing premise

The preference pairs and chosen diversity metric correctly identify sequences that are both diverse and structurally consistent for peptides, and OpenFold-generated structures serve as reliable conditioning inputs.

What would settle it

An evaluation on held-out peptide structures showing that the fine-tuned sequences produce no measurable gain in structural similarity scores or sequence diversity relative to the base ProteinMPNN or standard DPO outputs.

Figures

Figures reproduced from arXiv: 2410.19471 by Bruno Trentini, C. Brian Roland, Chen Tessler, Darren J. Hsu, Maria Korshunova, Olivia Viessmann, Ryan Park, Shie Mannor.

**Figure 1.** Figure 1: Motivation for DPO design choices. Left. Frequency of amino acid across ProteinMPNN generations conditioned on the peptide train set, vs. frequency over the peptide sequences. Middle. When conditioned on the same structure, the diversity of sequences generated by base ProteinMPNN does not correlate with the diversity of sequences generated by fine-tuned ProteinMPNN. Right. Distribution of rank correlation … view at source ↗

**Figure 2.** Figure 2: Exploring the effect of online diversity optimization. [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗

**Figure 3.** Figure 3: Pareto front and KL divergences. Left. Pareto front for various α values over temperature sweep. Middle. KL divergence from πref for various α values. α = 0 has β = 0.5, all other have β = 0.1. Right. KL divergence for DPO with reward scaling (β = 0.1) and without (β = 0.5). We develop a new understanding of ProteinMPNN to explain this. ProteinMPNN performs randomorder token decoding during sampling and a… view at source ↗

**Figure 4.** Figure 4: Sequence recovery with diversity optimization. [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗

**Figure 5.** Figure 5: Reward scaling improves DPO. Left. Reward-scaled DPO is a Pareto improvement over standard DPO over a temperature sweep. Middle. Left axis is the KL divergence between the token frequencies in the peptide train set vs. model samples (lower is better), right axis is the fraction of non-repeating tokens (higher is better). Right. TM-score improvement over base DPO. Lower TMscore buckets contain structures f… view at source ↗

**Figure 6.** Figure 6: Sequence samples on three randomly chosen structures from the CATH 4.3 peptide benchmark, all [PITH_FULL_IMAGE:figures/full_fig_p017_6.png] view at source ↗

**Figure 7.** Figure 7: Sequence samples on three randomly chosen structures from the OpenFold peptide benchmark, all [PITH_FULL_IMAGE:figures/full_fig_p017_7.png] view at source ↗

read the original abstract

Inverse folding models play an important role in structure-based design by predicting amino acid sequences that fold into desired reference structures. Models like ProteinMPNN, a message-passing encoder-decoder model, are trained to reliably produce new sequences from a reference structure. However, when applied to peptides, these models are prone to generating repetitive sequences that do not fold into the reference structure. To address this, we fine-tune ProteinMPNN to produce diverse and structurally consistent peptide sequences via Direct Preference Optimization (DPO). We derive two enhancements to DPO: online diversity regularization and domain-specific priors. Additionally, we develop a new understanding on improving diversity in decoder models. When conditioned on OpenFold generated structures, our fine-tuned models achieve state-of-the-art structural similarity scores, improving base ProteinMPNN by at least 8%. Compared to standard DPO, our regularized method achieves up to 20% higher sequence diversity with no loss in structural similarity score.

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.

Desk Editor's Note private letter to a colleague

The paper adds online diversity regularization to DPO on ProteinMPNN for peptides and reports diversity gains, but the structural claims rest on OpenFold outputs whose reliability for short peptides is untested.

read the letter

The core move here is taking ProteinMPNN, generating preference pairs from it, and then fine-tuning with DPO plus two tweaks: an online diversity regularizer and some peptide-specific priors. The reported outcome is that the regularized version keeps structural similarity scores while lifting sequence diversity by up to 20% over plain DPO, and the whole thing beats the base model by at least 8% on structure when conditioned on OpenFold backbones. That is the actual new piece: a concrete adaptation of DPO for the repetition problem that shows up in peptide inverse folding. It is a narrow but practical step rather than a new algorithm family. The work is honest about the starting problem and tries to measure both diversity and consistency, which is better than many decoder papers that only report one or the other. The diversity regularization looks like a sensible way to push the model away from repetitive outputs without an extra loss term that has to be tuned from scratch. The soft spot is the evaluation backbone. All the gains are measured against OpenFold-generated structures, and the abstract gives no sign that any test cases were checked against experimental PDB entries or even short MD runs. OpenFold was trained on longer globular proteins, so its accuracy on flexible peptides is known to drop; if the preference pairs themselves were ranked by agreement with those same OpenFold outputs, the loop can reward sequences that fit the model's biases rather than real energetics. That makes the 8% and 20% numbers hard to interpret without more validation. The paper is aimed at groups already running ProteinMPNN on peptide design tasks who want a drop-in fine-tuning recipe. It is not broad enough to change how most people think about inverse folding in general. It deserves a serious referee because the method is reproducible in principle and the diversity claim is falsifiable, but the referee should be asked to check whether the structural metric holds up once real structures are substituted. I would send it to review with that caveat rather than desk-reject.

Referee Report

1 major / 0 minor

Summary. The manuscript claims that fine-tuning ProteinMPNN via Direct Preference Optimization augmented with online diversity regularization and domain-specific priors yields peptide sequences with improved structural consistency and diversity. Conditioned on OpenFold-generated structures, the approach reports at least 8% higher structural similarity than base ProteinMPNN and up to 20% higher sequence diversity than standard DPO with no loss in structural score.

Significance. If the empirical improvements prove robust under independent structural validation, the work would supply a practical recipe for mitigating repetitive outputs in inverse folding decoders applied to peptides, a setting where standard models underperform. The emphasis on decoder diversity regularization could also transfer to other sequence generation tasks.

major comments (1)

[Abstract / Evaluation] The central performance claims (Abstract) are conditioned exclusively on OpenFold-generated reference structures for both training preference pairs and evaluation metrics. No cross-validation against experimental PDB peptide entries or orthogonal folding simulations (e.g., MD) is reported, which is load-bearing: OpenFold was trained predominantly on longer globular proteins, and its accuracy on short, flexible peptides is known to degrade, creating a risk that the optimization loop rewards model-consistent rather than physically consistent sequences.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their thoughtful and constructive review. We address the major comment point-by-point below.

read point-by-point responses

Referee: [Abstract / Evaluation] The central performance claims (Abstract) are conditioned exclusively on OpenFold-generated reference structures for both training preference pairs and evaluation metrics. No cross-validation against experimental PDB peptide entries or orthogonal folding simulations (e.g., MD) is reported, which is load-bearing: OpenFold was trained predominantly on longer globular proteins, and its accuracy on short, flexible peptides is known to degrade, creating a risk that the optimization loop rewards model-consistent rather than physically consistent sequences.

Authors: We appreciate the referee highlighting this important consideration. Our work specifically evaluates improvements to inverse folding when conditioned on OpenFold-generated structures, which represents a common practical use case for peptide design where experimental structures are frequently unavailable. All reported gains (at least 8% structural similarity over base ProteinMPNN and up to 20% sequence diversity over standard DPO) are measured relative to the baseline under identical OpenFold conditioning, demonstrating that the proposed diversity-regularized DPO yields sequences that are both more diverse and more consistent with the provided reference structure. We agree that the absence of experimental PDB cross-validation or MD simulations leaves open the possibility that improvements partly reflect alignment to OpenFold's own biases rather than independent physical validity. In the revised manuscript we will add an expanded Limitations section that explicitly discusses reliance on computational structures, notes the known limitations of OpenFold on short peptides, and outlines the value of future experimental or MD-based validation. We do not claim the generated sequences are guaranteed to fold experimentally, only that they exhibit improved consistency with the reference model. revision: partial

Circularity Check

0 steps flagged

No circularity; empirical ML application with no self-referential derivations

full rationale

The paper describes an empirical fine-tuning procedure applying DPO (with proposed online diversity regularization and domain-specific priors) to ProteinMPNN, reporting measured gains in structural similarity and sequence diversity on OpenFold-conditioned peptide tasks. No equations, uniqueness theorems, or derivation chains appear in the abstract or described content. All central claims are framed as experimental outcomes against external metrics rather than reductions to fitted parameters or self-citations by construction. The work is therefore self-contained as a standard empirical study.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only view yields no explicit free parameters, axioms, or invented entities; the approach relies on standard DPO machinery and ProteinMPNN without new postulated objects.

pith-pipeline@v0.9.0 · 5717 in / 1023 out tokens · 21020 ms · 2026-05-23T19:01:54.443161+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

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

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

We derive two enhancements to DPO: online diversity regularization and domain-specific priors... LDPO (πθ; πref) = −E(x,yw,yl)∼D [log σ (β log πθ (yw | x)/πref (yw | x) − β log πθ (yl | x)/πref (yl | x) + αEy′∼π̃(y′ | x) [γ(yl, y′)] − αEy′∼π̃(y′ | x) [γ(yw, y′)])]
IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

When conditioned on OpenFold generated structures, our fine-tuned models achieve state-of-the-art structural similarity scores

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 3 Pith papers

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

Entropy Across the Bridge: Conditional-Marginal Discretization for Flow and Schr\"odinger Samplers
cs.LG 2026-05 unverdicted novelty 7.0

Derives a conditional-marginal entropy-rate objective for bridge-aware discretization that yields U-shaped schedules and improves low-NFE sample quality on 2D, CIFAR-10, and protein tasks.
Pushing Biomolecular Utility-Diversity Frontiers with Supergroup Relative Policy Optimization
cs.CE 2026-05 unverdicted novelty 6.0

SGRPO expands the utility-diversity Pareto frontier in biomolecular design by using supergroup sampling and leave-one-out diversity rewards combined with utility signals.
Pushing Biomolecular Utility-Diversity Frontiers with Supergroup Relative Policy Optimization
cs.CE 2026-05 conditional novelty 6.0

SGRPO is a GRPO-style framework that constructs set-level diversity rewards via supergroup sampling and leave-one-out redistribution to expand the utility-diversity Pareto frontier in biomolecular design tasks.

Reference graph

Works this paper leans on

48 extracted references · 48 canonical work pages · cited by 2 Pith papers · 2 internal anchors

[1]

Abascal and Lynne Regan

Nadia C. Abascal and Lynne Regan. The past, present and future of protein-based materials. Open Biology, 8 0 (10), October 2018. ISSN 2046-2441. doi:10.1098/rsob.180113. URL http://dx.doi.org/10.1098/rsob.180113

work page doi:10.1098/rsob.180113 2018
[2]

O pen F old: R etraining A lpha F old2 yields new insights into its learning mechanisms and capacity for generalization

Gustaf Ahdritz, Nazim Bouatta, Christina Floristean, Sachin Kadyan, Qinghui Xia, William Gerecke, Timothy J O Donnell, Daniel Berenberg, Ian Fisk, Niccolò Zanichelli, Bo Zhang, Arkadiusz Nowaczynski, Bei Wang, Marta M Stepniewska-Dziubinska, Shang Zhang, Adegoke Ojewole, Murat Efe Guney, Stella Biderman, Andrew M Watkins, Stephen Ra, Pablo Ribalta Lorenzo...

work page doi:10.1101/2022.11.20.517210 2022
[3]

Ralph Allan Bradley and Milton E. Terry. Rank analysis of incomplete block designs: I. the method of paired comparisons. Biometrika, 39 0 (3/4): 0 324--345, 1952. ISSN 00063444, 14643510. URL http://www.jstor.org/stable/2334029

work page arXiv 1952
[4]

Design and application of stimulus-responsive peptide systems

Karuppiah Chockalingam, Mark Blenner, and Scott Banta. Design and application of stimulus-responsive peptide systems. Protein Engineering, Design & Selection, 20 0 (4): 0 155--161, 2007

work page 2007
[5]

Cell-penetrating peptides: design, synthesis, and applications

Dana Maria Copolovici, Kent Langel, Elo Eriste, and Ulo Langel. Cell-penetrating peptides: design, synthesis, and applications. ACS nano, 8 0 (3): 0 1972--1994, 2014

work page 1972
[6]

Paul E. Correa. The building of protein structures from alpha-carbon coordinates. Proteins: Structure, Function, and Bioinformatics, 7 0 (4): 0 366--377, 1990. doi:https://doi.org/10.1002/prot.340070408. URL https://onlinelibrary.wiley.com/doi/abs/10.1002/prot.340070408

work page doi:10.1002/prot.340070408 1990
[7]

Dauparas, I

J. Dauparas, I. Anishchenko, N. Bennett, H. Bai, R. J. Ragotte, L. F. Milles, B. I. M. Wicky, A. Courbet, R. J. de Haas, N. Bethel, P. J. Y. Leung, T. F. Huddy, S. Pellock, D. Tischer, F. Chan, B. Koepnick, H. Nguyen, A. Kang, B. Sankaran, A. K. Bera, N. P. King, and D. Baker. Robust deep learning–based protein sequence design using proteinmpnn. Science, ...

work page doi:10.1126/science.add2187 2022
[8]

Robust deep learning--based protein sequence design using proteinmpnn

Justas Dauparas, Ivan Anishchenko, Nathaniel Bennett, Hua Bai, Robert J Ragotte, Lukas F Milles, Basile IM Wicky, Alexis Courbet, Rob J de Haas, Neville Bethel, et al. Robust deep learning--based protein sequence design using proteinmpnn. Science, 378 0 (6615): 0 49--56, 2022 b

work page 2022
[9]

Ken A. Dill, S. Banu Ozkan, M. Scott Shell, and Thomas R. Weikl. The protein folding problem. Annual Review of Biophysics, 37 0 (1): 0 289–316, June 2008. ISSN 1936-1238. doi:10.1146/annurev.biophys.37.092707.153558. URL http://dx.doi.org/10.1146/annurev.biophys.37.092707.153558

work page doi:10.1146/annurev.biophys.37.092707.153558 2008
[10]

Pifold: Toward effective and efficient protein inverse folding

Zhangyang Gao, Cheng Tan, Pablo Chac \'o n, and Stan Z Li. Pifold: Toward effective and efficient protein inverse folding. arXiv preprint arXiv:2209.12643, 2022

work page arXiv 2022
[11]

Zhangyang Gao, Cheng Tan, and Stan Z. Li. Knowledge-design: Pushing the limit of protein design via knowledge refinement, 2023 a . URL https://arxiv.org/abs/2305.15151

work page arXiv 2023
[12]

Zhangyang Gao, Cheng Tan, Yijie Zhang, Xingran Chen, Lirong Wu, and Stan Z. Li. Proteininvbench: Benchmarking protein inverse folding on diverse tasks, models, and metrics. In Thirty-seventh Conference on Neural Information Processing Systems Datasets and Benchmarks Track, 2023 b . URL https://openreview.net/forum?id=bqXduvuW5E

work page 2023
[13]

De novo and inverse folding predictions of protein structure and dynamics

A Godzik, A Kolinski, and J Skolnick. De novo and inverse folding predictions of protein structure and dynamics. J. Comput. Aided Mol. Des., 7 0 (4): 0 397--438, August 1993

work page 1993
[15]

Learning inverse folding from millions of predicted structures

Chloe Hsu, Robert Verkuil, Jason Liu, Zeming Lin, Brian Hie, Tom Sercu, Adam Lerer, and Alexander Rives. Learning inverse folding from millions of predicted structures. bioRxiv, 2022 b . doi:10.1101/2022.04.10.487779. URL https://www.biorxiv.org/content/early/2022/04/10/2022.04.10.487779

work page doi:10.1101/2022.04.10.487779 2022
[16]

Generative models for graph-based protein design

John Ingraham, Vikas Garg, Regina Barzilay, and Tommi Jaakkola. Generative models for graph-based protein design. In H. Wallach, H. Larochelle, A. Beygelzimer, F. d Alch\' e -Buc, E. Fox, and R. Garnett (eds.), Advances in Neural Information Processing Systems, volume 32. Curran Associates, Inc., 2019. URL https://proceedings.neurips.cc/paper_files/paper/...

work page 2019
[17]

Equivariant graph neural networks for 3d macromolecular structure

Bowen Jing, Stephan Eismann, Pratham N Soni, and Ron O Dror. Equivariant graph neural networks for 3d macromolecular structure. arXiv preprint arXiv:2106.03843, 2021 a

work page arXiv 2021
[18]

Learning from protein structure with geometric vector perceptrons

Bowen Jing, Stephan Eismann, Patricia Suriana, Raphael John Lamarre Townshend, and Ron Dror. Learning from protein structure with geometric vector perceptrons. In International Conference on Learning Representations, 2021 b . URL https://openreview.net/forum?id=1YLJDvSx6J4

work page 2021
[19]

John Jumper, Richard Evans, Alexander Pritzel, Tim Green, Michael Figurnov, Olaf Ronneberger, Kathryn Tunyasuvunakool, Russ Bates, Augustin Žídek, Anna Potapenko, Alex Bridgland, Clemens Meyer, Simon A. A. Kohl, Andrew J. Ballard, Andrew Cowie, Bernardino Romera-Paredes, Stanislav Nikolov, Rishub Jain, Jonas Adler, Trevor Back, Stig Petersen, David Reiman...

work page doi:10.1038/s41586-021-03819-2 2021
[20]

Adam: A method for stochastic optimization

Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization. 2014

work page 2014
[21]

De novo protein design

Patrice Koehl and Michael Levitt. De novo protein design. i. in search of stability and specificity11edited by f. e. cohen. Journal of Molecular Biology, 293 0 (5): 0 1161--1181, 1999. ISSN 0022-2836. doi:https://doi.org/10.1006/jmbi.1999.3211. URL https://www.sciencedirect.com/science/article/pii/S0022283699932114

work page doi:10.1006/jmbi.1999.3211 1999
[22]

and Leibler, R

Kullback, S. and Leibler, R. A. On information and sufficiency. The Annals of Mathematical Statistics, 22(1): 0 79--86., 1951

work page 1951
[23]

Macromolecular modeling and design in rosetta: recent methods and frameworks

Julia Koehler Leman, Brian D Weitzner, Steven M Lewis, Jared Adolf-Bryfogle, Nawsad Alam, Rebecca F Alford, Melanie Aprahamian, David Baker, Kyle A Barlow, Patrick Barth, Benjamin Basanta, Brian J Bender, Kristin Blacklock, Jaume Bonet, Scott E Boyken, Phil Bradley, Chris Bystroff, Patrick Conway, Seth Cooper, Bruno E Correia, Brian Coventry, Rhiju Das, R...

work page 2020
[24]

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, Allan dos Santos Costa, Maryam Fazel-Zarandi, Tom Sercu, Salvatore Candido, and Alexander Rives. Evolutionary-scale prediction of atomic level protein structure with a language model. July 2022. doi:10.1101/2022.07.20.500902...

work page doi:10.1101/2022.07.20.500902 2022
[25]

Predicting the conformations of peptides and proteins in early evolution

E James Milner-White and Michael J Russell. Predicting the conformations of peptides and proteins in early evolution. a review article submitted to biology direct. Biol. Direct, 3 0 (1): 0 3, January 2008

work page 2008
[26]

Colabfold: making protein folding accessible to all

Milot Mirdita, Konstantin Schütze, Yoshitaka Moriwaki, Lim Heo, Sergey Ovchinnikov, and Martin Steinegger. Colabfold: making protein folding accessible to all. Nature Methods, 19 0 (6): 0 679--682, June 2022. doi:10.1038/s41592-022-01488-1. URL https://doi.org/10.1038/s41592-022-01488-1

work page doi:10.1038/s41592-022-01488-1 2022
[27]

Preference optimization of protein language models as a multi-objective binder design paradigm, 2024

Pouria Mistani and Venkatesh Mysore. Preference optimization of protein language models as a multi-objective binder design paradigm, 2024. URL https://arxiv.org/abs/2403.04187

work page arXiv 2024
[28]

Long Ouyang, Jeff Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul Christiano, Jan Leike, and Ryan Lowe. Training language models to follow instructions with human feedback,...

work page internal anchor Pith review Pith/arXiv arXiv 2022
[29]

Preference Optimization for Molecular Language Models

Ryan Park, Ryan Theisen, Navriti Sahni, Marcel Patek, Anna Cichońska, and Rayees Rahman. Preference optimization for molecular language models, 2023. URL https://arxiv.org/abs/2310.12304

work page arXiv 2023
[30]

Disentangling length from quality in direct preference optimization, 2024

Ryan Park, Rafael Rafailov, Stefano Ermon, and Chelsea Finn. Disentangling length from quality in direct preference optimization, 2024. URL https://arxiv.org/abs/2403.19159

work page arXiv 2024
[31]

PyTorch : An imperative style, high-performance deep learning library

Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, Alban Desmaison, Andreas K \"o pf, Edward Yang, Zach DeVito, Martin Raison, Alykhan Tejani, Sasank Chilamkurthy, Benoit Steiner, Lu Fang, Junjie Bai, and Soumith Chintala. PyTorch : An imperative style, high-per...

work page 2019
[32]

Direct Preference Optimization: Your Language Model is Secretly a Reward Model

Rafael Rafailov, Archit Sharma, Eric Mitchell, Stefano Ermon, Christopher D. Manning, and Chelsea Finn. Direct preference optimization: Your language model is secretly a reward model. ArXiv, abs/2305.18290, 2023. URL https://api.semanticscholar.org/CorpusID:258959321

work page internal anchor Pith review Pith/arXiv arXiv 2023
[33]

Shanker, Theodora U.J

Varun R. Shanker, Theodora U.J. Bruun, Brian L. Hie, and Peter S. Kim. Inverse folding of protein complexes with a structure-informed language model enables unsupervised antibody evolution. December 2023. doi:10.1101/2023.12.19.572475. URL http://dx.doi.org/10.1101/2023.12.19.572475

work page doi:10.1101/2023.12.19.572475 2023
[34]

Mmseqs2 enables sensitive protein sequence searching for the analysis of massive data sets

Martin Steinegger and Johannes Soeding. Mmseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nature Biotechnology, 35 0 (11): 0 1026--1028, 2017. doi:10.1038/nbt.3988. URL https://doi.org/10.1038/nbt.3988

work page doi:10.1038/nbt.3988 2017
[35]

Axel Tiessen, Paulino P \'e rez-Rodr \' guez, and Luis Jos \'e Delaye-Arredondo. Mathematical modeling and comparison of protein size distribution in different plant, animal, fungal and microbial species reveals a negative correlation between protein size and protein number, thus providing insight into the evolution of proteomes. BMC Res. Notes, 5 0 (1): ...

work page 2012
[36]

Peptide design principles for antimicrobial applications

Marcelo DT Torres, Shanmugapriya Sothiselvam, Timothy K Lu, and Cesar de la Fuente-Nunez. Peptide design principles for antimicrobial applications. Journal of molecular biology, 431 0 (18): 0 3547--3567, 2019

work page 2019
[37]

Designing peptide based nanomaterials

Rein V Ulijn and Andrew M Smith. Designing peptide based nanomaterials. Chemical Society Reviews, 37 0 (4): 0 664--675, 2008

work page 2008
[38]

Beyond reverse kl: Generalizing direct preference optimization with diverse divergence constraints, 2023

Chaoqi Wang, Yibo Jiang, Chenghao Yang, Han Liu, and Yuxin Chen. Beyond reverse kl: Generalizing direct preference optimization with diverse divergence constraints, 2023. URL https://arxiv.org/abs/2309.16240

work page arXiv 2023
[39]

Aligning protein generative models with experimental fitness via direct preference optimization

Talal Widatalla, Rafael Rafailov, and Brian Hie. Aligning protein generative models with experimental fitness via direct preference optimization. bioRxiv, 2024. doi:10.1101/2024.05.20.595026. URL https://www.biorxiv.org/content/early/2024/05/21/2024.05.20.595026

work page doi:10.1101/2024.05.20.595026 2024
[40]

Alphafold2 and its applications in the fields of biology and medicine

Zhenyu Yang, Xiaoxi Zeng, Yi Zhao, and Runsheng Chen. Alphafold2 and its applications in the fields of biology and medicine. Signal Transduction and Targeted Therapy, 8 0 (1), March 2023. ISSN 2059-3635. doi:10.1038/s41392-023-01381-z. URL http://dx.doi.org/10.1038/s41392-023-01381-z

work page doi:10.1038/s41392-023-01381-z 2023
[41]

Graph denoising diffusion for inverse protein folding

Kai Yi, Bingxin Zhou, Yiqing Shen, Pietro Li \`o , and Yu Guang Wang. Graph denoising diffusion for inverse protein folding. 2023

work page 2023
[42]

Inverse protein folding problem: designing polymer sequences

Kaizhi Yue and Ken A Dill. Inverse protein folding problem: designing polymer sequences. Proceedings of the National Academy of Sciences, 89 0 (9): 0 4163--4167, 1992

work page 1992
[43]

Scoring function for automated assessment of protein structure template quality

Yang Zhang and Jeffrey Skolnick. Scoring function for automated assessment of protein structure template quality. Proteins: Structure, Function, and Bioinformatics, 57 0 (4): 0 702--710, 2004. ISSN 1097-0134. doi:10.1002/prot.20264. URL http://bioinformatics.buffalo.edu/TM-score. Copyright 2004 Wiley-Liss, Inc

work page doi:10.1002/prot.20264 2004
[44]

Beyond one-preference-fits-all alignment: Multi-objective direct preference optimization, 2024

Zhanhui Zhou, Jie Liu, Jing Shao, Xiangyu Yue, Chao Yang, Wanli Ouyang, and Yu Qiao. Beyond one-preference-fits-all alignment: Multi-objective direct preference optimization, 2024. URL https://arxiv.org/abs/2310.03708

work page arXiv 2024
[45]

Ziebart, Andrew Maas, J

Brian D. Ziebart, Andrew Maas, J. Andrew Bagnell, and Anind K. Dey. Maximum entropy inverse reinforcement learning. In Proceedings of the 23rd National Conference on Artificial Intelligence - Volume 3, AAAI'08, pp.\ 1433–1438. AAAI Press, 2008. ISBN 9781577353683

work page 2008
[46]

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...

work page
[47]

@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 not add it explicitly Type <Return> for now, but then later remove the command n...

work page
[48]

\@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@first@sw \@firstoftwo \@ifundefined NAT@b*@#2 \@firstoftwo @num @NAT@ctr \@secondoft...

work page
[49]

@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 \@bibsetup #1 @NAT@ctr @ @openbib .11em \@plus.33em \@minus.07em 4000 4000 `\.\@m @bibit...

work page

[1] [1]

Abascal and Lynne Regan

Nadia C. Abascal and Lynne Regan. The past, present and future of protein-based materials. Open Biology, 8 0 (10), October 2018. ISSN 2046-2441. doi:10.1098/rsob.180113. URL http://dx.doi.org/10.1098/rsob.180113

work page doi:10.1098/rsob.180113 2018

[2] [2]

O pen F old: R etraining A lpha F old2 yields new insights into its learning mechanisms and capacity for generalization

Gustaf Ahdritz, Nazim Bouatta, Christina Floristean, Sachin Kadyan, Qinghui Xia, William Gerecke, Timothy J O Donnell, Daniel Berenberg, Ian Fisk, Niccolò Zanichelli, Bo Zhang, Arkadiusz Nowaczynski, Bei Wang, Marta M Stepniewska-Dziubinska, Shang Zhang, Adegoke Ojewole, Murat Efe Guney, Stella Biderman, Andrew M Watkins, Stephen Ra, Pablo Ribalta Lorenzo...

work page doi:10.1101/2022.11.20.517210 2022

[3] [3]

Ralph Allan Bradley and Milton E. Terry. Rank analysis of incomplete block designs: I. the method of paired comparisons. Biometrika, 39 0 (3/4): 0 324--345, 1952. ISSN 00063444, 14643510. URL http://www.jstor.org/stable/2334029

work page arXiv 1952

[4] [4]

Design and application of stimulus-responsive peptide systems

Karuppiah Chockalingam, Mark Blenner, and Scott Banta. Design and application of stimulus-responsive peptide systems. Protein Engineering, Design & Selection, 20 0 (4): 0 155--161, 2007

work page 2007

[5] [5]

Cell-penetrating peptides: design, synthesis, and applications

Dana Maria Copolovici, Kent Langel, Elo Eriste, and Ulo Langel. Cell-penetrating peptides: design, synthesis, and applications. ACS nano, 8 0 (3): 0 1972--1994, 2014

work page 1972

[6] [6]

Paul E. Correa. The building of protein structures from alpha-carbon coordinates. Proteins: Structure, Function, and Bioinformatics, 7 0 (4): 0 366--377, 1990. doi:https://doi.org/10.1002/prot.340070408. URL https://onlinelibrary.wiley.com/doi/abs/10.1002/prot.340070408

work page doi:10.1002/prot.340070408 1990

[7] [7]

Dauparas, I

J. Dauparas, I. Anishchenko, N. Bennett, H. Bai, R. J. Ragotte, L. F. Milles, B. I. M. Wicky, A. Courbet, R. J. de Haas, N. Bethel, P. J. Y. Leung, T. F. Huddy, S. Pellock, D. Tischer, F. Chan, B. Koepnick, H. Nguyen, A. Kang, B. Sankaran, A. K. Bera, N. P. King, and D. Baker. Robust deep learning–based protein sequence design using proteinmpnn. Science, ...

work page doi:10.1126/science.add2187 2022

[8] [8]

Robust deep learning--based protein sequence design using proteinmpnn

Justas Dauparas, Ivan Anishchenko, Nathaniel Bennett, Hua Bai, Robert J Ragotte, Lukas F Milles, Basile IM Wicky, Alexis Courbet, Rob J de Haas, Neville Bethel, et al. Robust deep learning--based protein sequence design using proteinmpnn. Science, 378 0 (6615): 0 49--56, 2022 b

work page 2022

[9] [9]

Ken A. Dill, S. Banu Ozkan, M. Scott Shell, and Thomas R. Weikl. The protein folding problem. Annual Review of Biophysics, 37 0 (1): 0 289–316, June 2008. ISSN 1936-1238. doi:10.1146/annurev.biophys.37.092707.153558. URL http://dx.doi.org/10.1146/annurev.biophys.37.092707.153558

work page doi:10.1146/annurev.biophys.37.092707.153558 2008

[10] [10]

Pifold: Toward effective and efficient protein inverse folding

Zhangyang Gao, Cheng Tan, Pablo Chac \'o n, and Stan Z Li. Pifold: Toward effective and efficient protein inverse folding. arXiv preprint arXiv:2209.12643, 2022

work page arXiv 2022

[11] [11]

Zhangyang Gao, Cheng Tan, and Stan Z. Li. Knowledge-design: Pushing the limit of protein design via knowledge refinement, 2023 a . URL https://arxiv.org/abs/2305.15151

work page arXiv 2023

[12] [12]

Zhangyang Gao, Cheng Tan, Yijie Zhang, Xingran Chen, Lirong Wu, and Stan Z. Li. Proteininvbench: Benchmarking protein inverse folding on diverse tasks, models, and metrics. In Thirty-seventh Conference on Neural Information Processing Systems Datasets and Benchmarks Track, 2023 b . URL https://openreview.net/forum?id=bqXduvuW5E

work page 2023

[13] [13]

De novo and inverse folding predictions of protein structure and dynamics

A Godzik, A Kolinski, and J Skolnick. De novo and inverse folding predictions of protein structure and dynamics. J. Comput. Aided Mol. Des., 7 0 (4): 0 397--438, August 1993

work page 1993

[14] [15]

Learning inverse folding from millions of predicted structures

Chloe Hsu, Robert Verkuil, Jason Liu, Zeming Lin, Brian Hie, Tom Sercu, Adam Lerer, and Alexander Rives. Learning inverse folding from millions of predicted structures. bioRxiv, 2022 b . doi:10.1101/2022.04.10.487779. URL https://www.biorxiv.org/content/early/2022/04/10/2022.04.10.487779

work page doi:10.1101/2022.04.10.487779 2022

[15] [16]

Generative models for graph-based protein design

John Ingraham, Vikas Garg, Regina Barzilay, and Tommi Jaakkola. Generative models for graph-based protein design. In H. Wallach, H. Larochelle, A. Beygelzimer, F. d Alch\' e -Buc, E. Fox, and R. Garnett (eds.), Advances in Neural Information Processing Systems, volume 32. Curran Associates, Inc., 2019. URL https://proceedings.neurips.cc/paper_files/paper/...

work page 2019

[16] [17]

Equivariant graph neural networks for 3d macromolecular structure

Bowen Jing, Stephan Eismann, Pratham N Soni, and Ron O Dror. Equivariant graph neural networks for 3d macromolecular structure. arXiv preprint arXiv:2106.03843, 2021 a

work page arXiv 2021

[17] [18]

Learning from protein structure with geometric vector perceptrons

Bowen Jing, Stephan Eismann, Patricia Suriana, Raphael John Lamarre Townshend, and Ron Dror. Learning from protein structure with geometric vector perceptrons. In International Conference on Learning Representations, 2021 b . URL https://openreview.net/forum?id=1YLJDvSx6J4

work page 2021

[18] [19]

John Jumper, Richard Evans, Alexander Pritzel, Tim Green, Michael Figurnov, Olaf Ronneberger, Kathryn Tunyasuvunakool, Russ Bates, Augustin Žídek, Anna Potapenko, Alex Bridgland, Clemens Meyer, Simon A. A. Kohl, Andrew J. Ballard, Andrew Cowie, Bernardino Romera-Paredes, Stanislav Nikolov, Rishub Jain, Jonas Adler, Trevor Back, Stig Petersen, David Reiman...

work page doi:10.1038/s41586-021-03819-2 2021

[19] [20]

Adam: A method for stochastic optimization

Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization. 2014

work page 2014

[20] [21]

De novo protein design

Patrice Koehl and Michael Levitt. De novo protein design. i. in search of stability and specificity11edited by f. e. cohen. Journal of Molecular Biology, 293 0 (5): 0 1161--1181, 1999. ISSN 0022-2836. doi:https://doi.org/10.1006/jmbi.1999.3211. URL https://www.sciencedirect.com/science/article/pii/S0022283699932114

work page doi:10.1006/jmbi.1999.3211 1999

[21] [22]

and Leibler, R

Kullback, S. and Leibler, R. A. On information and sufficiency. The Annals of Mathematical Statistics, 22(1): 0 79--86., 1951

work page 1951

[22] [23]

Macromolecular modeling and design in rosetta: recent methods and frameworks

Julia Koehler Leman, Brian D Weitzner, Steven M Lewis, Jared Adolf-Bryfogle, Nawsad Alam, Rebecca F Alford, Melanie Aprahamian, David Baker, Kyle A Barlow, Patrick Barth, Benjamin Basanta, Brian J Bender, Kristin Blacklock, Jaume Bonet, Scott E Boyken, Phil Bradley, Chris Bystroff, Patrick Conway, Seth Cooper, Bruno E Correia, Brian Coventry, Rhiju Das, R...

work page 2020

[23] [24]

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, Allan dos Santos Costa, Maryam Fazel-Zarandi, Tom Sercu, Salvatore Candido, and Alexander Rives. Evolutionary-scale prediction of atomic level protein structure with a language model. July 2022. doi:10.1101/2022.07.20.500902...

work page doi:10.1101/2022.07.20.500902 2022

[24] [25]

Predicting the conformations of peptides and proteins in early evolution

E James Milner-White and Michael J Russell. Predicting the conformations of peptides and proteins in early evolution. a review article submitted to biology direct. Biol. Direct, 3 0 (1): 0 3, January 2008

work page 2008

[25] [26]

Colabfold: making protein folding accessible to all

Milot Mirdita, Konstantin Schütze, Yoshitaka Moriwaki, Lim Heo, Sergey Ovchinnikov, and Martin Steinegger. Colabfold: making protein folding accessible to all. Nature Methods, 19 0 (6): 0 679--682, June 2022. doi:10.1038/s41592-022-01488-1. URL https://doi.org/10.1038/s41592-022-01488-1

work page doi:10.1038/s41592-022-01488-1 2022

[26] [27]

Preference optimization of protein language models as a multi-objective binder design paradigm, 2024

Pouria Mistani and Venkatesh Mysore. Preference optimization of protein language models as a multi-objective binder design paradigm, 2024. URL https://arxiv.org/abs/2403.04187

work page arXiv 2024

[27] [28]

Long Ouyang, Jeff Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul Christiano, Jan Leike, and Ryan Lowe. Training language models to follow instructions with human feedback,...

work page internal anchor Pith review Pith/arXiv arXiv 2022

[28] [29]

Preference Optimization for Molecular Language Models

Ryan Park, Ryan Theisen, Navriti Sahni, Marcel Patek, Anna Cichońska, and Rayees Rahman. Preference optimization for molecular language models, 2023. URL https://arxiv.org/abs/2310.12304

work page arXiv 2023

[29] [30]

Disentangling length from quality in direct preference optimization, 2024

Ryan Park, Rafael Rafailov, Stefano Ermon, and Chelsea Finn. Disentangling length from quality in direct preference optimization, 2024. URL https://arxiv.org/abs/2403.19159

work page arXiv 2024

[30] [31]

PyTorch : An imperative style, high-performance deep learning library

Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, Alban Desmaison, Andreas K \"o pf, Edward Yang, Zach DeVito, Martin Raison, Alykhan Tejani, Sasank Chilamkurthy, Benoit Steiner, Lu Fang, Junjie Bai, and Soumith Chintala. PyTorch : An imperative style, high-per...

work page 2019

[31] [32]

Direct Preference Optimization: Your Language Model is Secretly a Reward Model

Rafael Rafailov, Archit Sharma, Eric Mitchell, Stefano Ermon, Christopher D. Manning, and Chelsea Finn. Direct preference optimization: Your language model is secretly a reward model. ArXiv, abs/2305.18290, 2023. URL https://api.semanticscholar.org/CorpusID:258959321

work page internal anchor Pith review Pith/arXiv arXiv 2023

[32] [33]

Shanker, Theodora U.J

Varun R. Shanker, Theodora U.J. Bruun, Brian L. Hie, and Peter S. Kim. Inverse folding of protein complexes with a structure-informed language model enables unsupervised antibody evolution. December 2023. doi:10.1101/2023.12.19.572475. URL http://dx.doi.org/10.1101/2023.12.19.572475

work page doi:10.1101/2023.12.19.572475 2023

[33] [34]

Mmseqs2 enables sensitive protein sequence searching for the analysis of massive data sets

Martin Steinegger and Johannes Soeding. Mmseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nature Biotechnology, 35 0 (11): 0 1026--1028, 2017. doi:10.1038/nbt.3988. URL https://doi.org/10.1038/nbt.3988

work page doi:10.1038/nbt.3988 2017

[34] [35]

Axel Tiessen, Paulino P \'e rez-Rodr \' guez, and Luis Jos \'e Delaye-Arredondo. Mathematical modeling and comparison of protein size distribution in different plant, animal, fungal and microbial species reveals a negative correlation between protein size and protein number, thus providing insight into the evolution of proteomes. BMC Res. Notes, 5 0 (1): ...

work page 2012

[35] [36]

Peptide design principles for antimicrobial applications

Marcelo DT Torres, Shanmugapriya Sothiselvam, Timothy K Lu, and Cesar de la Fuente-Nunez. Peptide design principles for antimicrobial applications. Journal of molecular biology, 431 0 (18): 0 3547--3567, 2019

work page 2019

[36] [37]

Designing peptide based nanomaterials

Rein V Ulijn and Andrew M Smith. Designing peptide based nanomaterials. Chemical Society Reviews, 37 0 (4): 0 664--675, 2008

work page 2008

[37] [38]

Beyond reverse kl: Generalizing direct preference optimization with diverse divergence constraints, 2023

Chaoqi Wang, Yibo Jiang, Chenghao Yang, Han Liu, and Yuxin Chen. Beyond reverse kl: Generalizing direct preference optimization with diverse divergence constraints, 2023. URL https://arxiv.org/abs/2309.16240

work page arXiv 2023

[38] [39]

Aligning protein generative models with experimental fitness via direct preference optimization

Talal Widatalla, Rafael Rafailov, and Brian Hie. Aligning protein generative models with experimental fitness via direct preference optimization. bioRxiv, 2024. doi:10.1101/2024.05.20.595026. URL https://www.biorxiv.org/content/early/2024/05/21/2024.05.20.595026

work page doi:10.1101/2024.05.20.595026 2024

[39] [40]

Alphafold2 and its applications in the fields of biology and medicine

Zhenyu Yang, Xiaoxi Zeng, Yi Zhao, and Runsheng Chen. Alphafold2 and its applications in the fields of biology and medicine. Signal Transduction and Targeted Therapy, 8 0 (1), March 2023. ISSN 2059-3635. doi:10.1038/s41392-023-01381-z. URL http://dx.doi.org/10.1038/s41392-023-01381-z

work page doi:10.1038/s41392-023-01381-z 2023

[40] [41]

Graph denoising diffusion for inverse protein folding

Kai Yi, Bingxin Zhou, Yiqing Shen, Pietro Li \`o , and Yu Guang Wang. Graph denoising diffusion for inverse protein folding. 2023

work page 2023

[41] [42]

Inverse protein folding problem: designing polymer sequences

Kaizhi Yue and Ken A Dill. Inverse protein folding problem: designing polymer sequences. Proceedings of the National Academy of Sciences, 89 0 (9): 0 4163--4167, 1992

work page 1992

[42] [43]

Scoring function for automated assessment of protein structure template quality

Yang Zhang and Jeffrey Skolnick. Scoring function for automated assessment of protein structure template quality. Proteins: Structure, Function, and Bioinformatics, 57 0 (4): 0 702--710, 2004. ISSN 1097-0134. doi:10.1002/prot.20264. URL http://bioinformatics.buffalo.edu/TM-score. Copyright 2004 Wiley-Liss, Inc

work page doi:10.1002/prot.20264 2004

[43] [44]

Beyond one-preference-fits-all alignment: Multi-objective direct preference optimization, 2024

Zhanhui Zhou, Jie Liu, Jing Shao, Xiangyu Yue, Chao Yang, Wanli Ouyang, and Yu Qiao. Beyond one-preference-fits-all alignment: Multi-objective direct preference optimization, 2024. URL https://arxiv.org/abs/2310.03708

work page arXiv 2024

[44] [45]

Ziebart, Andrew Maas, J

Brian D. Ziebart, Andrew Maas, J. Andrew Bagnell, and Anind K. Dey. Maximum entropy inverse reinforcement learning. In Proceedings of the 23rd National Conference on Artificial Intelligence - Volume 3, AAAI'08, pp.\ 1433–1438. AAAI Press, 2008. ISBN 9781577353683

work page 2008

[45] [46]

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...

work page

[46] [47]

@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 not add it explicitly Type <Return> for now, but then later remove the command n...

work page

[47] [48]

\@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@first@sw \@firstoftwo \@ifundefined NAT@b*@#2 \@firstoftwo @num @NAT@ctr \@secondoft...

work page

[48] [49]

@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 \@bibsetup #1 @NAT@ctr @ @openbib .11em \@plus.33em \@minus.07em 4000 4000 `\.\@m @bibit...

work page