Paper Citation Record · LEDGER

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction

As of 10 August 2026, this Paper Citation Record lists 38 of 38 outbound references and 0 inbound Pith citation observations for arXiv:2608.06582.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2608.06582 v1

Coverage vector

measured 38 of 38 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-10T04:23:54.881662Z

measured 38 of 38 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-10T06:31:04.303077+00:00

measured 0 of 0 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: cited_works

Reference resolution

38 of 38 outbound references displayed

verified exact1
verified fuzzy27
unresolved8
parse uncertain0
malformed identifier2
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 17b784ae-dc61-43f3-b349-4ddb7a407a88 · outbound

This paper cites Crystal structure generation with autoregressive large lan- guage modeling.Nature Communications, 15(1):10570,.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Crystal structure generation with autoregressive large lan- guage modeling.Nature Communications, 15(1):10570,

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.515824Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation d81cfd15-9157-4061-91a5-7437e03e087f · outbound

This paper cites A foundation model for atomistic materials chemistry.The Journal of chemical physics, 163(18), 2025.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction A foundation model for atomistic materials chemistry.The Journal of chemical physics, 163(18), 2025

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.504452Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 9bbf9776-c120-4cda-9075-6eea90afd06a · outbound

This paper cites Training Diffusion Models with Reinforcement Learning.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Training Diffusion Models with Reinforcement Learning

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-10T04:23:54.646822Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ebb4992a-6385-4aac-8bb8-fe6e97fa60ef · outbound

This paper cites Matinvent: Reinforcement learning for 3d crystal diffusion generation.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Matinvent: Reinforcement learning for 3d crystal diffusion generation

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.492053Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation a4e9cb6e-51db-4516-97ed-e82317016c0c · outbound

This paper cites Fine-tuned language models generate stable inorganic ma- terials as text.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Fine-tuned language models generate stable inorganic ma- terials as text

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.459297Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation bed0d57c-6ce5-4b99-9b84-8776c601e967 · outbound

This paper cites Open materials generation with inference-time reinforcement learning.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Open materials generation with inference-time reinforcement learning

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.447750Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation c47e4555-02d8-49b8-a561-1fe5614b3c60 · outbound

This paper cites Tadmor, and Stefano Martiniani.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Tadmor, and Stefano Martiniani

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.437379Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 0b57a060-a1d2-4ff8-b8cd-ca3d54125fa7 · outbound

This paper cites Crystal structure prediction by joint equivariant diffusion.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Crystal structure prediction by joint equivariant diffusion

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.427267Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 11280999-1397-4eaf-8da4-56a484a2743a · outbound

This paper cites Space Group Constrained Crystal Generation.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Space Group Constrained Crystal Generation

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-10T04:23:54.717071Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4a0b3320-cc5d-4838-8d9b-db6cf586a6c6 · outbound

This paper cites Wyckoffdiff–a generative diffusion model for crystal symmetry.arXiv preprint arXiv:2502.06485, 2025.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Wyckoffdiff–a generative diffusion model for crystal symmetry.arXiv preprint arXiv:2502.06485, 2025

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-10T04:23:54.720950Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c61185e3-3802-41e8-97e2-a7c34105a85c · outbound

This paper cites Llm meets dif- fusion: a hybrid framework for crystal material generation.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Llm meets dif- fusion: a hybrid framework for crystal material generation

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.399011Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 1abca9a6-07f0-4954-b957-02dd011bc379 · outbound

This paper cites SymmCD: Symmetry- preserving crystal generation with diffusion models.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction SymmCD: Symmetry- preserving crystal generation with diffusion models

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.382682Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation cb898a7e-58c0-4a9f-948b-155040794ec8 · outbound

This paper cites Powder diffraction crystal structure determination us- ing generative models.Nature Communications, 16(1):7428,.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Powder diffraction crystal structure determination us- ing generative models.Nature Communications, 16(1):7428,

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.365490Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation a2593708-18c8-4765-bfe8-9ec21b63d9d4 · outbound

This paper cites Flow matching for generative modeling.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Flow matching for generative modeling

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.332855Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 04ff9146-7a43-4041-8974-a55288505888 · outbound

This paper cites Flow-GRPO: Training flow matching models via online RL.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Flow-GRPO: Training flow matching models via online RL

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.321630Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation ef4532db-2896-42e5-89a1-13348acb2ea2 · outbound

This paper cites Crystalflow: a flow-based generative model for crystalline materials.Nature Communications, 16(1):9267, 2025.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Crystalflow: a flow-based generative model for crystalline materials.Nature Communications, 16(1):9267, 2025

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.311393Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 139722fa-609e-4b3c-b5c2-7f54fc2028bc · outbound

This paper cites All that structure matches does not glitter.Advances in Neural Information Processing Systems, 38, 2026.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction All that structure matches does not glitter.Advances in Neural Information Processing Systems, 38, 2026

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.299988Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 854ff117-f71b-456e-826f-9b0b4c6975ee · outbound

This paper cites Genera- tive models for crystalline materials.Advanced Materials, 38(18):e23620, 2026.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Genera- tive models for crystalline materials.Advanced Materials, 38(18):e23620, 2026

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.289550Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation bb60e780-07da-4435-a81f-a3c55f162463 · outbound

This paper cites an unresolved cited work.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Unresolved cited work

Reference 19

Resolution: unresolved; raw_fallback, observed 2026-08-10T04:23:55.279640Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 5a1fa568-f54e-4616-b3a3-72a20b6c9a40 · outbound

This paper cites Reliable crystal structure predictions from first principles.Nature Commu- nications, 13(1):3095, 2022.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Reliable crystal structure predictions from first principles.Nature Commu- nications, 13(1):3095, 2022

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.269070Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 7054eed9-58cd-47a6-8291-86455421d8e7 · outbound

This paper cites Python materials genomics (pymatgen): A robust, open-source python library for materials analysis.Computa- tional Materials Science, 68:314–319, 2013.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Python materials genomics (pymatgen): A robust, open-source python library for materials analysis.Computa- tional Materials Science, 68:314–319, 2013

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.255391Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation e48ee655-b63e-4c27-925e-8931533132e2 · outbound

This paper cites Guiding generative mod- els to uncover diverse and novel crystals via reinforcement learning.Nature Machine Intelligence, pages 1–13, 2026.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Guiding generative mod- els to uncover diverse and novel crystals via reinforcement learning.Nature Machine Intelligence, pages 1–13, 2026

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.238629Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 0d637131-48c8-412d-b24f-53edf4837533 · outbound

This paper cites Ai- assisted rapid crystal structure generation towards a target local environment.npj Computational Materials, 2026.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Ai- assisted rapid crystal structure generation towards a target local environment.npj Computational Materials, 2026

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.225927Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation e7bef633-bcb9-448d-8595-8b4e310dea1b · outbound

This paper cites Proximal Policy Optimization Algorithms.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Proximal Policy Optimization Algorithms

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-10T04:23:54.823746Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a11357a2-ffc6-47f5-8b2d-cb3376fcfefa · outbound

This paper cites DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-10T04:23:54.828112Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3403ee94-1699-40c6-8edf-bd109577ca05 · outbound

This paper cites Jaakkola, Elsa Olivetti, and Rafael G ´omez-Bombarelli.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Jaakkola, Elsa Olivetti, and Rafael G ´omez-Bombarelli

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.213781Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 90917ed7-dc12-44f8-bc39-6fdfa89abb1b · outbound

This paper cites The thermo- dynamic scale of inorganic crystalline metastability.Science advances, 2(11):e1600225, 2016.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction The thermo- dynamic scale of inorganic crystalline metastability.Science advances, 2(11):e1600225, 2016

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.200470Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 16c6db0d-ca04-4022-96b2-351604c17d3c · outbound

This paper cites Fine-Tuning of Continuous-Time Diffusion Models as Entropy-Regularized Control.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Fine-Tuning of Continuous-Time Diffusion Models as Entropy-Regularized Control

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-10T04:23:54.840347Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5b339336-e94f-4a90-9587-d281755b003e · outbound

This paper cites Pass@k policy optimization: Solving harder reinforcement learning prob- lems.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Pass@k policy optimization: Solving harder reinforcement learning prob- lems

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.182967Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation eaeda7ed-e156-44ff-8767-f3f2c30560be · outbound

This paper cites A Periodic Bayesian Flow for Material Generation.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction A Periodic Bayesian Flow for Material Generation

Reference 30

Resolution: verified exact; local_arxiv, observed 2026-08-10T04:23:54.918452Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation df90b34c-ac09-459c-b3d0-d3fa5450a8b7 · outbound

This paper cites Jaakkola.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Jaakkola

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.170733Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation dc9b1723-a21b-45ab-be15-05405bfb26d6 · outbound

This paper cites Does rein- forcement learning really incentivize reasoning capacity in LLMs beyond the base model? InThe Thirty-ninth An- nual Conference on Neural Information Processing Systems,.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Does rein- forcement learning really incentivize reasoning capacity in LLMs beyond the base model? InThe Thirty-ninth An- nual Conference on Neural Information Processing Systems,

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.159204Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation b3ffbc10-5c18-4a1d-94da-4017cbe0f1ad · outbound

This paper cites A generative model for inorganic materials design.Nature, 2025.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction A generative model for inorganic materials design.Nature, 2025

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.148152Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 2d42b46e-afd8-4477-b448-1c63275c1833 · outbound

This paper cites an unresolved cited work.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Unresolved cited work

Reference 34

Resolution: unresolved; raw_fallback, observed 2026-08-10T04:23:55.136221Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 03c017e6-55a7-464f-b928-411c68013926 · outbound

This paper cites an unresolved cited work.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Unresolved cited work

Reference 35

Resolution: malformed identifier; raw_fallback, observed 2026-08-10T04:23:55.125155Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation ffcd3e68-46a5-4a4b-b204-466be0991306 · outbound

This paper cites Both variants follow the same on-policy GRPO pipeline, alternating between stochastic rollout collection and policy optimization.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction Both variants follow the same on-policy GRPO pipeline, alternating between stochastic rollout collection and policy optimization

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.112120Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 0e9e5549-2b69-4978-b0cc-9949812842a4 · outbound

This paper cites CSP-aligned Hybrid Reward 3.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction CSP-aligned Hybrid Reward 3

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-08-10T04:23:55.099476Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 6a6c5ff2-43c8-4c76-be62-05549f0bc0cb · outbound

This paper cites CSP-aligned Hybrid Reward For completeness and reproducibility, we provide the ex- act reward implementation used to obtain the reported re- sults.

CrystalGRPO: Target-Aligned and Coverage-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction CSP-aligned Hybrid Reward For completeness and reproducibility, we provide the ex- act reward implementation used to obtain the reported re- sults

Reference 38

Resolution: malformed identifier; raw_fallback, observed 2026-08-10T04:23:55.085852Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Pith citing papers

No inbound Pith citation observations are available.