Polymer Data Challenges in the AI Era: Bridging Gaps for Next-Generation Energy Materials

Guanhua Chen; Jie Liu; Ying Zhao

REVIEW 3 major objections 4 minor 18 references

Polymer Data Challenges in the AI Era: Bridging Gaps for Next-Generation Energy Materials

T0 review · 3 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read This review claims that polymer data fragmentation, inconsistent testing standards, and incomplete metadata—not algorithm design—are the main bottleneck for AI-driven discovery of energy polymers, and that FAIR-compliant data…

desk verdict A useful map of polymer data resources, but the specific statistics are often misattributed or unverifiable—worth peer review after major citation surgery. read the letter →

arxiv 2505.13494 v1 pith:ZM33RFQI submitted 2025-05-15 cond-mat.soft cond-mat.mtrl-scics.LG

Ying Zhao , Guanhua Chen , Jie Liu This is my paper · ORCID

classification cond-mat.softcond-mat.mtrl-scics.LG

keywords polymerinformaticsFAIRdataprinciplesmachinelearningformaterialsenergynaturallanguageprocessingautonomousexperimentationstandardizationdatabases

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

This review argues that the main barrier to using machine learning to design better polymers for batteries, solar cells, and hydrogen storage is not the algorithms but the data. It claims polymer results are scattered across disconnected papers and proprietary repositories, that fewer than 15 percent of published polymer studies give machine-readable synthesis details, and that inconsistent testing methods make the same material look different from one laboratory to another. The paper identifies three systemic barriers—academic–industrial data silos, inconsistent measurement protocols, and incomplete metadata—and argues that fixing them, through data principles requiring results to be findable, accessible, interoperable, and reusable, together with text-mining tools and automated laboratories, would accelerate energy-materials discovery. A sympathetic reader takes away that data infrastructure, not model architecture, is the binding constraint on AI-driven polymer discovery.

What carries the argument

The central object is the FAIR-compliant polymer data record: every measured property—ionic conductivity, glass-transition temperature, viscosity, dielectric strength—must be accompanied by machine-readable metadata describing synthesis, processing, characterization, and environmental conditions. The argument is carried by pairing this record with two generation engines. One is natural-language processing, including large language models, that extracts structured entries from papers, patents, and legacy figures. The other is the closed-loop autonomous laboratory, where machine-learning models select experiments and robotic platforms execute and characterize them, generating self-consistent data with recorded processing history. These components are meant to work together: literature mining fills historical gaps, autonomous labs create high-quality new data, and FAIR standards make both machine-actionable.

What would settle it

Check each headline figure against its cited source—for instance, whether reference [22] actually reports polyethylene oxide viscosity varying by 35 percent between academic and industrial measurements, and whether reference [35] actually reports a 62 percent drop in metadata inconsistencies. If one of these numbers cannot be found or refers to a different quantity, the review's central evidence fails.

Watch

Extended reading notes

Core claim

The paper's central claim is that polymer science is in a data crisis that is institutional and cultural as much as technical: without interoperable databases, machine-readable synthesis details, or standardized measurement protocols, cross-study comparison is unreliable and machine-learning models are biased toward a few well-documented polymers. The review's positive claim is that the crisis is solvable by three converging developments: FAIR-compliant polymer-specific ontologies, natural-language-processing tools that convert decades of unstructured literature into structured records, and high-throughput robotic platforms that produce self-consistent datasets through closed-loop experimentation. It also argues that cultural shifts toward open science and decentralized data sharing are necessary complements to these technologies. If the diagnosis is correct, the fastest route to better predictive models for polymer electrolytes, photovoltaics, and hydrogen-storage membranes runs through data standardization and sharing rather than through further algorithmic improvements alone.

Load-bearing premise

The review's case rests on the accuracy and correct attribution of its headline statistics; if the reported percentages do not actually appear in the cited sources or measure something else, the evidence for a distinct polymer data crisis is not established.

Editorial extensions

If this is right

If data fragmentation is the binding constraint, standardizing reporting and metadata should improve machine-learning accuracy on polymer properties more than further model innovations do.
Universal testing protocols—for example, specifying humidity during impedance spectroscopy—would reduce cross-laboratory discrepancies such as the 35 percent viscosity spread cited for polyethylene oxide.
NLP-driven extraction could convert decades of polymer literature into training data, expanding coverage to understudied families like vitrimers and conjugated microporous polymers.
Autonomous laboratories could produce self-consistent datasets hundreds of times faster than manual synthesis, enabling models to be trained on data with complete processing histories.
Pre-competitive data pools and federated sharing would let industrial processing data contribute to models without exposing proprietary formulations.

Reading between the lines

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

A direct consequence the review leaves implicit is that today's model-comparison benchmarks in polymer informatics may be measuring data quality rather than model quality; standardizing data could reshuffle those comparisons.
The same recipe—FAIR metadata, literature mining, and closed-loop experimentation—appears transferable to other soft-matter and formulation sciences, not only energy polymers.
A testable extension would be a controlled benchmark in which the same models are trained on legacy data versus FAIR-complete data with identical chemical coverage, isolating the contribution of metadata completeness to prediction error.
The review's argument implies that publishers, funders, and industrial consortia hold the main levers; a randomized evaluation of journal data mandates would provide causal evidence for that claim.

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.

Desk Editor's Note

read the letter

This is a survey that will help someone entering polymer informatics, but you shouldn't quote any of its specific statistics without checking the original sources.

The paper does a decent job of mapping the landscape. It covers PolyInfo, NanoMine, Polymer Genome, Khazana, the major NLP tools (SciBERT, MatBERT, PolyBERT), high-throughput platforms, autonomous laboratories, and crowdsourced databases in one place. The organization is logical, and the core observation—that polymer data are fragmented and that FAIR, NLP, and automated experimentation are part of the solution—matches what I know of the field. For a newcomer who wants a list of resources and methods, this is a reasonable starting point.

The problem is the evidence. The review leans on a set of numbers that are either misattributed or untraceable. The 35% viscosity spread for PEO is credited to a nanomedicine methods paper that has nothing to do with polymer electrolytes. The 'over 30%' predictive-accuracy loss is credited to two NRC reports from 1999 and 2008 that don't contain that analysis. The MaTCH paper is cited for a 62% reduction in metadata inconsistencies—that figure is not in that paper. ACS Macro Letters is credited with a 55% drop in reproducibility claims, with no source at all. The Polymer Data Alliance and Polymer Metadata Consortium, the IUPAC 2024 FAIR mandate, and NIST's blockchain certification appear without identifiable references. These aren't side details; they are what the 'data crisis' argument is built on. If the numbers can't be traced, the review is an opinion piece with citations fixed to it.

The paper introduces no new data, method, or theory, which is fine for a review, but it means the synthesis has to be accurate, and this one isn't yet.

I'd send this to peer review rather than desk-reject it, because the topic is important and the authors have assembled a broad, useful overview. But the referees should be told to check every quantitative claim and every institution named. The authors should be asked to remove or properly source the unverifiable figures and organizations before publication. After that, it could serve as a helpful orientation for students and researchers new to polymer data.

Referee Report

3 major / 4 minor

Summary. The manuscript is a perspective/review arguing that fragmented, non-standardized polymer data ecosystems constitute the primary bottleneck for AI-driven polymer discovery, particularly for energy materials. It catalogs five systemic barriers (data fragmentation, metadata crisis, measurement anarchy, industrial-academic divide, and lack of cohesion) and surveys emerging solutions including NLP-based literature mining, high-throughput experimentation, autonomous laboratories, FAIR principles adapted to polymer-specific ontologies, crowdsourced databases, and decentralized data governance. The authors advocate for a multi-pronged technical and cultural shift toward interoperable, machine-readable polymer data as an essential enabler for next-generation energy materials.

Significance. If its claims were properly sourced, the paper would offer a useful synthesis of the current state and future directions of polymer informatics, with particular value in connecting data curation, ML, experimental automation, and policy. Its breadth is a strength, and the authors correctly emphasize that algorithmic advances alone are insufficient without better data infrastructure. However, the paper's evidentiary foundation is compromised by numerous misattributed or unverifiable quantitative statistics, and the central argument that data fragmentation is the primary bottleneck rests heavily on those figures. The potential contribution is real, but the manuscript in its present form cannot be relied upon as an evidence-based assessment; it requires substantial revision to correct or remove unsupported claims.

major comments (3)

[§2.3] The claim that "viscosity data for PEO varied by 35% between academic publications and industrial technical datasheets" (paragraph 3) is attributed to ref. 22, Sharifi et al., Nano-Micro Lett. 2022, which is a paper on standardizing analytical characterization in nanomedicine literature and contains no such PEO viscosity statistic. Similarly, the assertion that "fewer than 20% of studies on PEO-based electrolytes specify the moisture levels" is attributed to ref. 23, Albright and Chai, Environ. Sci. Technol. 2021, a paper on polymer biodegradation knowledge gaps. These misattributions are load-bearing because they constitute the quantitative evidence for the "measurement anarchy" claimed in this section. The authors should either supply correct sources for these numbers or remove them.
[§2.4] The statement that ML models trained on academic data "fail to account for industrial-scale variables... a gap that reduces predictive accuracy by over 30% when validated against manufacturing datasets" (paragraph 2) cites refs. 29 and 30, which are National Research Council reports on industrial technology assessments and integrated computational materials engineering; neither reports an ML validation study nor a 30% figure. Without a traceable source, this statistic cannot support the "data chasm" argument. The accompanying assertion that joint industry-academia projects "stall" due to IP disputes is also cited to the same unrelated NRC reports.
[§2.5 (and §2.3)] This section introduces multiple quantitative and institutional claims with inadequate or absent sourcing: the "62% reduction in metadata inconsistencies" attributed to MaTCH (ref. 35), the "55% reduction in data irreproducibility claims" attributed to ACS Macro Letters, the "IUPAC 2024 FAIR mandate," the "NIST blockchain-based certification systems," and the "Polymer Data Alliance" and "Polymer Metadata Consortium" (the latter first appears in §2.3). The cited references (refs. 33, 35, 36) are about machine learning in polymer composites, environmental microplastics, and life-cycle impacts, respectively, and do not support these specific entities or metrics. These claims must be verified, properly referenced, or removed; otherwise the paper presents unsupported "facts" as the basis for its recommendations.

minor comments (4)

[Header] The word "Aknowledgement" should be corrected to "Acknowledgement."
[§1 and §3.3] The in-text citation "Jurğis6 et al." is incorrect; reference 6 is by Ruza et al., so the text should say "Ruza et al." This error appears both in the Introduction and in Section 3.3.
[§2.1] The sentence "the study by Geiculescu10 demonstrated... However, these simulations often rely on simplified models" is confusing because reference 10 is an experimental study on PEG plasticizers, not a simulation study. The sentence should be rewritten to clarify which simulations are being critiqued.
[Reference list] Several references appear mismatched with the citations in the text (e.g., ref. 44 is listed as "Polymer Property Predictor and Database. NIST." but the text refers to it as "NIST Polymer Database"). A full audit of the reference list against the in-text citations is needed.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper is a literature review with no derived claims, fitted parameters, or load-bearing self-citations.

full rationale

This manuscript is a review/position paper, not a derivation-driven study. It makes no mathematical claims, fits no parameters, and presents no predictions that could reduce to its own inputs. Its central assertion—that fragmented, non-standardized polymer data hinders AI-driven energy-materials discovery—is supported by citations to external literature (e.g., Shetty et al., Polymer Genome, MaTCH, PolyInfo) rather than by any self-referential chain. None of the authors' own prior work is invoked as load-bearing evidence, so there is no self-citation circularity. The skeptic's concern that several quantitative statistics (e.g., the 35% PEO viscosity variation, the 62% metadata-inconsistency reduction) appear misattributed or untraceable in the cited references is a factual/correctness critique about evidence quality, not a circularity argument; per the review rules, that concern belongs under correctness risk rather than circularity scoring. The paper does not present a derivation whose conclusion is equivalent to its assumptions by construction, nor does it rename a known empirical pattern as a new result. Honest non-finding is therefore appropriate.

Assumptions & free parameters 0 free parameters · 3 assumptions · 2 invented entities

No free parameters are present because the paper reports no new measurements or fits. The axioms are the review's unverified factual dependencies. Two entities are flagged as invented because the paper presents them as existing organizations without any citable source.

assumptions (3)

domain assumption The cited statistics and facts are accurately represented from the named sources.
The review's factual claims rest on these citations; several are mis-attributed (e.g., refs 22, 23, 29, 30, 35), so this assumption is questionable.
domain assumption The organizations 'Polymer Data Alliance' and 'Polymer Metadata Consortium' exist and operate as described.
No reference is provided for either organization; the cited works do not mention them, so this is an unverified existential claim.
standard math Background domain knowledge (FAIR principles, DFT/MD simulation, PSMILES, battery/electrolyte terminology) is understood by the intended audience.
The review relies on standard concepts without explanation; this is normal for a review and not a weakness.

invented entities (2)

Polymer Metadata Consortium
purpose: Introduced as a blockchain-based crowdsourcing platform for polymer protocol evaluation (Section 2.3).
No citation supports its existence; the paper's cited references do not describe such a consortium.
Polymer Data Alliance
purpose: Introduced as a consortium of 32 academic and industrial partners running a pre-competitive data pool (Section 2.5).
No citation supports its existence; cited refs 33 and 36 do not mention this alliance.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Polymer Data Challenges in the AI Era: Bridging Gaps for Next-Generation Energy Materials." pith.science (2026). https://pith.science/paper/ZM33RFQI

@misc{pith2026250513494,
  author       = {Pith},
  title        = {Pith review of: Polymer Data Challenges in the AI Era: Bridging Gaps for Next-Generation Energy Materials},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZM33RFQI}},
  note         = {Machine review of arXiv:2505.13494}
}

read the original abstract

The pursuit of advanced polymers for energy technologies, spanning photovoltaics, solid-state batteries, and hydrogen storage, is hindered by fragmented data ecosystems that fail to capture the hierarchical complexity of these materials. Polymer science lacks interoperable databases, forcing reliance on disconnected literature and legacy records riddled with unstructured formats and irreproducible testing protocols. This fragmentation stifles machine learning (ML) applications and delays the discovery of materials critical for global decarbonization. Three systemic barriers compound the challenge. First, academic-industrial data silos restrict access to proprietary industrial datasets, while academic publications often omit critical synthesis details. Second, inconsistent testing methods undermine cross-study comparability. Third, incomplete metadata in existing databases limits their utility for training reliable ML models. Emerging solutions address these gaps through technological and collaborative innovation. Natural language processing (NLP) tools extract structured polymer data from decades of literature, while high-throughput robotic platforms generate self-consistent datasets via autonomous experimentation. Central to these advances is the adoption of FAIR (Findable, Accessible, Interoperable, Reusable) principles, adapted to polymer-specific ontologies, ensuring machine-readability and reproducibility. Future breakthroughs hinge on cultural shifts toward open science, accelerated by decentralized data markets and autonomous laboratories that merge robotic experimentation with real-time ML validation. By addressing data fragmentation through technological innovation, collaborative governance, and ethical stewardship, the polymer community can transform bottlenecks into accelerants.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

18 extracted references · 12 canonical work pages

[1]

room temperature

figures, demanding sophisticated curation strategies to transform fragmented insights into computable datasets. Traditional manual extraction methods, while reliable, face insurmountable scalability challenges in this era of big data. For instance, 1210 experimentally measured permittivity values belong to 738 unique polymers were manually collected from ...

work page 2000
[4]

data divide

Future Directions and Outlook The rapid evolution of polymer science, driven by the urgent demand for advanced energy materials, necessitates a paradigm shift in how data ecosystems are designed, governed, and utilized. While significant progress has been made in addressing data fragmentation, quality, and standardization challenges, the road ahead requir...

2025
[6]

(34) Jayaraman, A.; Olsen, B

https://www.taylorfrancis.com/chapters/edit/10.1201/9781003476337-6/machine-learning-artificial-intelligence-polymer-composites-rajesh-mahadeva-sushanta-sethi-atul-maurya-saurav-dixit-vinay-gupta-gaurav-manik?locale=en (accessed 2025-05-14). (34) Jayaraman, A.; Olsen, B. Convergence of Artificial Intelligence, Machine Learning, Cheminformatics, and Polyme...

work page doi:10.1201/9781003476337-6/machine-learning-artificial-intelligence-polymer-composites-rajesh-mahadeva-sushanta-sethi-atul-maurya-saurav-dixit-vinay-gupta-gaurav-manik
[9]

Regimes of Ordering: The Commercialization of Intellectual Property in Industrial-Academic Collaborations

(31) Rappert, B.; and Webster, A. Regimes of Ordering: The Commercialization of Intellectual Property in Industrial-Academic Collaborations. Technol. Anal. Strateg. Manag. 1997, 9 (2), 115–130. https://doi.org/10.1080/09537329708524274. (32) Ge, W.; De Silva, R.; Fan, Y.; Sisson, S. A.; Stenzel, M. H. Machine Learning in Polymer Research. Adv. Mater. 2025...

work page doi:10.1080/09537329708524274 1997
[11]

38 (40) Otsuka, S.; Kuwajima, I.; Hosoya, J.; Xu, Y.; Yamazaki, M

https://doi.org/10.48550/arXiv.2302.13425. 38 (40) Otsuka, S.; Kuwajima, I.; Hosoya, J.; Xu, Y.; Yamazaki, M. PoLyInfo: Polymer Database for Polymeric Materials Design. In 2011 International Conference on Emerging Intelligent Data and Web Technologies; IEEE: Tirana, Albania, 2011; pp 22–29. https://doi.org/10.1109/EIDWT.2011.13. (41) Ma, R.; Zhang, H.; Xu...

work page doi:10.48550/arxiv.2302.13425 2011
[15]

From Tokens to Materials: Leveraging Language Models for Scientific Discovery

https://doi.org/10.48550/arXiv.2410.16165. (78) Kuenneth, C.; Ramprasad, R. polyBERT: A Chemical Language Model to Enable Fully Machine-Driven Ultrafast Polymer Informatics. Nat. Commun. 2023, 14 (1),

work page Pith review arXiv doi:10.48550/arxiv.2410.16165 2023
[172]

(23) Albright, V

https://doi.org/10.1007/s40820-022-00922-5. (23) Albright, V. C. I.; Chai, Y. Knowledge Gaps in Polymer Biodegradation Research. Environ. Sci. Technol. 2021, 55 (17), 11476–11488. https://doi.org/10.1021/acs.est.1c00994. (24) Sharifi, S.; Reuel, N.; Kallmyer, N.; Sun, E.; Landry, M. P.; Mahmoudi, M. The Issue of Reliability and Repeatability of Analytical...

work page arXiv 2021
[342]

(50) BiG-MAP: an Automated Pipeline To Profile Metabolic Gene Cluster Abundance and Expression in Microbiomes | mSystems

https://doi.org/10.1149/MA2024-023342mtgabs. (50) BiG-MAP: an Automated Pipeline To Profile Metabolic Gene Cluster Abundance and Expression in Microbiomes | mSystems. https://journals.asm.org/doi/full/10.1128/msystems.00937-21 (accessed 2025-05-14). (51) Home | NREL. https://www.nrel.gov/ (accessed 2025-05-14). (52) Materials Project. Materials Project. h...

work page doi:10.1149/ma2024-023342mtgabs

Show all 18 references

[1359]

(21) Tanaka, T

https://doi.org/10.3390/polym14071359. (21) Tanaka, T. Experimental Methods in Polymer Science: Modern Methods in Polymer Research and Technology; Elsevier,

doi
[1427]

(111) Shen, Z.-H.; Bao, Z.-W.; Cheng, X.-X.; Li, B.-W.; Liu, H.-X.; Shen, Y.; Chen, L.-Q.; Li, X.-G.; Nan, C.-W

https://doi.org/10.1088/0022-3727/39/7/014. (111) Shen, Z.-H.; Bao, Z.-W.; Cheng, X.-X.; Li, B.-W.; Liu, H.-X.; Shen, Y.; Chen, L.-Q.; Li, X.-G.; Nan, C.-W. Designing Polymer Nanocomposites with High Energy Density Using Machine Learning. Npj Comput. Mater. 2021, 7 (1), 1–9. h...

doi
[2008]

(20) Pires, J. R. A.; Souza, V. G. L.; Fuciños, P.; Pastrana, L.; Fernando, A. L. Methodologies to Assess the Biodegradability of Bio-Based Polymers—Current Knowledge and Existing Gaps. Polymers 2022, 14 (7),

2022
[2009]

Biomedical Named Entity Recognition Using Conditional Random Fields and Rich Feature Sets

(71) Settles, B. Biomedical Named Entity Recognition Using Conditional Random Fields and Rich Feature Sets. In Proceedings of the International Joint Workshop on Natural Language Processing in Biomedicine and its Applications (NLPBA/BioNLP); Collier, N., Ruch, P., Nazarenko, A...

2004
[2012]

N.; Voke, E.; Landry, M

(22) Sharifi, S.; Mahmoud, N. N.; Voke, E.; Landry, M. P.; Mahmoudi, M. Importance of Standardizing Analytical Characterization Methodology for Improved Reliability of the Nanomedicine Literature. Nano-Micro Lett. 2022, 14 (1),

2022
[2019]

(77) Wan, Y.; Xie, T.; Wu, N.; Zhang, W.; Kit, C.; Hoex, B

https://doi.org/10.48550/arXiv.1903.10676. (77) Wan, Y.; Xie, T.; Wu, N.; Zhang, W.; Kit, C.; Hoex, B. From Tokens to Materials: Leveraging Language Models for Scientific Discovery. arXiv November 3,

1903 doi arXiv
[2020]

(80) Afzal, M

https://doi.org/10.48550/arXiv.2010.09885. (80) Afzal, M. A. F.; Browning, A. R.; Goldberg, A.; Halls, M. D.; Gavartin, J. L.; Morisato, T.; Hughes, T. F.; Giesen, D. J.; Goose, J. E. High-Throughput Molecular Dynamics Simulations and Validation of Thermophysical Properties of...

2010 doi arXiv
[2024]

D.; Das, D.; Ramprasad, R

(16) Kim, C.; Chandrasekaran, A.; Huan, T. D.; Das, D.; Ramprasad, R. Polymer Genome: A Data-Powered Polymer Informatics Platform for Property Predictions. J. Phys. Chem. C 2018, 122 (31), 17575–17585. https://doi.org/10.1021/acs.jpcc.8b02913. (17) Chandrasekaran, A.; Kim, C.;...

2018 doi
[2025]

(7) Wilkinson, M

https://doi.org/10.26434/chemrxiv-2025-2cjbg. (7) Wilkinson, M. D.; Dumontier, M.; Aalbersberg, Ij. J.; Appleton, G.; Axton, M.; Baak, A.; Blomberg, N.; Boiten, J.-W.; da Silva Santos, L. B.; Bourne, P. E.; Bouwman, J.; Brookes, A. J.; Clark, T.; Crosas, M.; Dillo, I.; Dumon, ...

2025
[4099]

(79) Chithrananda, S.; Grand, G.; Ramsundar, B

https://doi.org/10.1038/s41467-023-39868-6. (79) Chithrananda, S.; Grand, G.; Ramsundar, B. ChemBERTa: Large-Scale Self-Supervised Pretraining for Molecular Property Prediction. arXiv October 23,

doi

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

room temperature

figures, demanding sophisticated curation strategies to transform fragmented insights into computable datasets. Traditional manual extraction methods, while reliable, face insurmountable scalability challenges in this era of big data. For instance, 1210 experimentally measured permittivity values belong to 738 unique polymers were manually collected from ...

work page 2000

[2] [4]

data divide

Future Directions and Outlook The rapid evolution of polymer science, driven by the urgent demand for advanced energy materials, necessitates a paradigm shift in how data ecosystems are designed, governed, and utilized. While significant progress has been made in addressing data fragmentation, quality, and standardization challenges, the road ahead requir...

2025

[3] [6]

(34) Jayaraman, A.; Olsen, B

https://www.taylorfrancis.com/chapters/edit/10.1201/9781003476337-6/machine-learning-artificial-intelligence-polymer-composites-rajesh-mahadeva-sushanta-sethi-atul-maurya-saurav-dixit-vinay-gupta-gaurav-manik?locale=en (accessed 2025-05-14). (34) Jayaraman, A.; Olsen, B. Convergence of Artificial Intelligence, Machine Learning, Cheminformatics, and Polyme...

work page doi:10.1201/9781003476337-6/machine-learning-artificial-intelligence-polymer-composites-rajesh-mahadeva-sushanta-sethi-atul-maurya-saurav-dixit-vinay-gupta-gaurav-manik

[4] [9]

Regimes of Ordering: The Commercialization of Intellectual Property in Industrial-Academic Collaborations

(31) Rappert, B.; and Webster, A. Regimes of Ordering: The Commercialization of Intellectual Property in Industrial-Academic Collaborations. Technol. Anal. Strateg. Manag. 1997, 9 (2), 115–130. https://doi.org/10.1080/09537329708524274. (32) Ge, W.; De Silva, R.; Fan, Y.; Sisson, S. A.; Stenzel, M. H. Machine Learning in Polymer Research. Adv. Mater. 2025...

work page doi:10.1080/09537329708524274 1997

[5] [11]

38 (40) Otsuka, S.; Kuwajima, I.; Hosoya, J.; Xu, Y.; Yamazaki, M

https://doi.org/10.48550/arXiv.2302.13425. 38 (40) Otsuka, S.; Kuwajima, I.; Hosoya, J.; Xu, Y.; Yamazaki, M. PoLyInfo: Polymer Database for Polymeric Materials Design. In 2011 International Conference on Emerging Intelligent Data and Web Technologies; IEEE: Tirana, Albania, 2011; pp 22–29. https://doi.org/10.1109/EIDWT.2011.13. (41) Ma, R.; Zhang, H.; Xu...

work page doi:10.48550/arxiv.2302.13425 2011

[6] [15]

From Tokens to Materials: Leveraging Language Models for Scientific Discovery

https://doi.org/10.48550/arXiv.2410.16165. (78) Kuenneth, C.; Ramprasad, R. polyBERT: A Chemical Language Model to Enable Fully Machine-Driven Ultrafast Polymer Informatics. Nat. Commun. 2023, 14 (1),

work page Pith review arXiv doi:10.48550/arxiv.2410.16165 2023

[7] [172]

(23) Albright, V

https://doi.org/10.1007/s40820-022-00922-5. (23) Albright, V. C. I.; Chai, Y. Knowledge Gaps in Polymer Biodegradation Research. Environ. Sci. Technol. 2021, 55 (17), 11476–11488. https://doi.org/10.1021/acs.est.1c00994. (24) Sharifi, S.; Reuel, N.; Kallmyer, N.; Sun, E.; Landry, M. P.; Mahmoudi, M. The Issue of Reliability and Repeatability of Analytical...

work page arXiv 2021

[8] [342]

(50) BiG-MAP: an Automated Pipeline To Profile Metabolic Gene Cluster Abundance and Expression in Microbiomes | mSystems

https://doi.org/10.1149/MA2024-023342mtgabs. (50) BiG-MAP: an Automated Pipeline To Profile Metabolic Gene Cluster Abundance and Expression in Microbiomes | mSystems. https://journals.asm.org/doi/full/10.1128/msystems.00937-21 (accessed 2025-05-14). (51) Home | NREL. https://www.nrel.gov/ (accessed 2025-05-14). (52) Materials Project. Materials Project. h...

work page doi:10.1149/ma2024-023342mtgabs