A Robust and Versatile Generative Model for Inverse Design of Polymers
Bibliographic record
Abstract
Efficiently designing polymers to meet specific requirements can expedite their translation into practical applications and lower development costs. Although generative inverse design is more efficient than trial-and-error or forward prediction–screening strategies, the imperfect validity of current polymer generative models prevents their seamless integration into scientific discovery workflows. Moreover, their limited controllability—such as the inability to reliably generate polymers with specific functional groups or classes—further constrains their practical utility. In this work, we integrate the robust Group SELFIES method with the state-of-the-art polymer generator PolyTAO to achieve generating 100% chemically valid polymer structures, removing a longstanding bottleneck in polymer design. Compared with previous polymer generation models, this work can generate—on demand—polymers that match specified chemical motifs, polymer classes, and target properties across an effectively unbounded chemical space. We further introduce a task-agnostic, continuous pretraining strategy that combines physics-informed heuristics with reinforcement learning. This approach preserves strong generative performance on user-defined tasks, even in low-data regimes. As a proof of concept, we rigorously validated the dielectric constants of 30 polyimides generated via controlled, on-demand design using first-principles calculations, finding deviations of less than 10% from their target values. Designed as a powerful backend engine for polymer inverse design, our model is deployment-ready, and integrates seamlessly with high-throughput, self-driving laboratories and industrial synthesis pipelines.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".