Papers › Crystal Diffusion Variational Autoencoder for Periodic Material Generation

Crystal Diffusion Variational Autoencoder for Periodic Material Generation

12 Oct 2021ICLR 2022 4arXiv:2110.06197archive 2025-07-28

Tian Xie, Xiang Fu, Octavian-Eugen Ganea, Regina Barzilay, Tommi Jaakkola

Generating the periodic structure of stable materials is a long-standing challenge for the material design community. This task is difficult because stable materials only exist in a low-dimensional subspace of all possible periodic arrangements of atoms: 1) the coordinates must lie in the local energy minimum defined by quantum mechanics, and 2) global stability also requires the structure to follow the complex, yet specific bonding preferences between different atom types. Existing methods fail to incorporate these factors and often lack proper invariances. We propose a Crystal Diffusion Variational Autoencoder (CDVAE) that captures the physical inductive bias of material stability. By learning from the data distribution of stable materials, the decoder generates materials in a diffusion process that moves atomic coordinates towards a lower energy state and updates atom types to satisfy bonding preferences between neighbors. Our model also explicitly encodes interactions across periodic boundaries and respects permutation, translation, rotation, and periodic invariances. We significantly outperform past methods in three tasks: 1) reconstructing the input structure, 2) generating valid, diverse, and realistic materials, and 3) generating materials that optimize a specific property. We also provide several standard datasets and evaluation metrics for the broader machine learning community.

PaperPDFConference PDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Syntology holds a reference implementation of Crystal Diffusion Variational Autoencoder (CDVAE), generated from the paper text (not the authors' code), at level 3 on Syntology's ladder: V2 “two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs” “neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)”; V3 “paper-derived property tests pass on both of those same two generations.” (quoted from Syntology's server card). No rung compares this code against the authors' implementation. For agents: get_reference_implementation(name='Crystal Diffusion Variational Autoencoder (CDVAE)') (how to connect); the free check have('2110.06197') needs no token.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 36 code samples harvested from 7 repositories linked to this paper: 27 ran (of which 0 constructed an object rather than computing a result; 22 with no instrument failure: 0 honoured, 0 violated, 22 with no contract checked; 5 where Syntology's instrument failed) · 9 unverified. Where Syntology's instrument failed: 5 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 2 samples from 1 repository, 1 ran (0 with no instrument failure; 1 where Syntology's instrument failed); community (archive-listed): 29 samples from 3 repositories, 22 ran (22 with no instrument failure; 0 where Syntology's instrument failed); community: 5 samples from 3 repositories, 4 ran (0 with no instrument failure; 4 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 2 of its repositories that Syntology harvested for other papers:

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 5 of the 27 samples that ran have no recorded date.

txie-93/cdvae officialmentioned in papermentioned on GitHubpytorchSyntology: 2 samples, 1 ran (1 instrument-failed) · 1 unverified report
cyye001/con-cdvae mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Con-CDVAE: A method for the conditional generation of crystal structures (arXiv:2403.12478; that paper's own run record: 6 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
facebookresearch/flowmm mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Analog Bits: Generating Discrete Data using Diffusion Models with Self-Conditioning (arXiv:2208.04202; that paper's own run record: 13 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 3 violated, 0 with no contract checked; 9 where Syntology's instrument failed) · 7 unverified) and Flow Matching on General Geometries (arXiv:2302.03660; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
knc6/cdvae mentioned on GitHubpytorchMITSyntology: 2 samples, 2 ran (0 instrument-failed) · 0 unverified report
sparks-baird/mp-time-split mentioned on GitHubMITSyntology: 3 samples, 2 ran (0 instrument-failed) · 1 unverified report
symmetryadvantage/wyckofftransformer mentioned on GitHubpytorchApache-2.0Syntology: 24 samples, 18 ran (0 instrument-failed) · 6 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

36 samples harvested: 27 ran (of which 0 constructed an object rather than computing a result; 22 with no instrument failure: 0 honoured, 0 violated, 22 with no contract checked; 5 where Syntology's instrument failed) · 9 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

5ran · our draft was wrong
22ran
9unverified

Licence: 0 of the 36 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 7 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

txie-93/cdvae official repository 2 samples, 1 ran (1 instrument-failed) · 1 unverified
generation in 9 papers scripts/evaluate.py ran · our draft was wrong MIT (permissive) · 2c52dde33acf320d · report fixture: signature model:callable/0/?, ld_kwargs:LDKwargs/0/?, num_batches_to_sample:int/0/int, num_samples_per_z:int/0/int, batch_size:int/0/int, down_sample_traj_step:int/0/int · returned tuple · level recorded, date not recorded
optimization scripts/evaluate.py unverified MIT (permissive) · 7d1d1bcb15f1fefe · report
knc6/cdvae community (archive-listed) 2 samples, 2 ran (0 instrument-failed) · 0 unverified
abs_cap cdvae/common/data_utils.py ran fingerprinted MIT (permissive) · 6a946510479f2069 · report fixture: signature val:float/0/float, max_abs_val:float/0/float · returned float · level recorded 2026-09-24
get_env in 13 papers cdvae/common/utils.py ran MIT (permissive) · 00e20379657c554c · report fixture: signature env_name:str/0/str, default:NoneType/0/? · returned str · level recorded 2026-09-24
sparks-baird/mp-time-split community (archive-listed) 3 samples, 2 ran (0 instrument-failed) · 1 unverified
fib src/mp_time_split/core.py ran fingerprinted MIT (permissive) · 23b0f488ae6bc68e · report fixture: signature n:int/0/int · returned int · level recorded 2026-09-24
get_data_home src/mp_time_split/core.py ran fingerprinted MIT (permissive) · d48fe570d3e210ff · report fixture: signature data_home:str/0/str · returned str · level recorded 2026-09-24
mp_time_split src/mp_time_split/utils/split.py unverified MIT (permissive) · 3bc0f40b2f7fd1cc · report
symmetryadvantage/wyckofftransformer community (archive-listed) 24 samples, 18 ran (0 instrument-failed) · 6 unverified
censored_min_mae src/wyckoff_transformer/censored.py ran Apache-2.0 (permissive) · 32b725dbd955ba54 · report fixture: signature prediction:Tensor/2/float, target:Tensor/1/float, criterion:callable/0/? · returned Tensor · level recorded 2026-09-24
censored_min_nll src/wyckoff_transformer/censored.py ran Apache-2.0 (permissive) · 31604f9db40cac45 · report fixture: signature observed:Tensor/2/float, location:Tensor/2/float, log_scale:Tensor/2/float, noise:float/0/float, min_scale:float/0/float · returned Tensor · level recorded 2026-09-24
chemical_system_conditioning_dim src/wyckoff_transformer/chemical_system.py ran fingerprinted Apache-2.0 (permissive) · 52d4978b57a845eb · report fixture: signature n_elements:int/0/int · returned int · level recorded 2026-09-24
chemical_system_vector src/wyckoff_transformer/chemical_system.py ran Apache-2.0 (permissive) · f1aa000edc093144 · report fixture: signature element_tokens:Tensor/1/int, n_elements:int/0/int, device:NoneType/0/? · returned Tensor · level recorded 2026-09-24
chemical_system_vectors src/wyckoff_transformer/chemical_system.py ran Apache-2.0 (permissive) · f5091d700a375822 · report fixture: signature element_tokens:list/1/list, n_elements:int/0/int, device:NoneType/0/? · returned Tensor · level recorded 2026-09-24
composition_conditioning_dim src/wyckoff_transformer/composition.py ran fingerprinted Apache-2.0 (permissive) · c4974f8a6be0acfa · report fixture: signature n_elements:int/0/int, condition_on_cell_size:bool/0/bool · returned int · level recorded 2026-09-24
composition_vector src/wyckoff_transformer/composition.py ran Apache-2.0 (permissive) · 3afdb8687fc3cf98 · report fixture: signature element_tokens:Tensor/1/int, counts:Tensor/1/float, n_elements:int/0/int, device:NoneType/0/?, condition_on_cell_size:bool/0/bool · returned Tensor · level recorded 2026-09-24
composition_vectors src/wyckoff_transformer/composition.py ran Apache-2.0 (permissive) · 5a3de7c294a992b4 · report fixture: signature element_tokens:list/1/list, counts:list/1/list, n_elements:int/0/int, device:device/0/?, condition_on_cell_size:bool/0/bool · returned Tensor · level recorded 2026-09-24
expected_min_bias src/wyckoff_transformer/censored.py ran fingerprinted Apache-2.0 (permissive) · 8818ca4313431c50 · report fixture: signature scale:Tensor/1/float, n:Tensor/1/float · returned Tensor · level recorded 2026-09-24
gene_representations src/wyckoff_transformer/gene_likelihood.py ran Apache-2.0 (permissive) · 70341785f156077f · report fixture: signature record:dict/0/? · returned tuple · level recorded 2026-09-24
group_multiplicity_db src/wyckoff_transformer/csp.py ran Apache-2.0 (permissive) · 03f72260a9ea750c · report fixture: signature multiplicity_db:dict/0/? · returned dict · level recorded 2026-09-24
log_orderings src/wyckoff_transformer/gene_likelihood.py ran fingerprinted Apache-2.0 (permissive) · a9dc0dc92f699021 · report fixture: signature sites:list/1/tuple · returned float · level recorded 2026-09-24
read_config in 49 papers src/wyckoff_transformer/paths.py ran Apache-2.0 (permissive) · 278184a2cd6bed62 · report fixture: signature path:PosixPath/0/? · returned dict · level recorded 2026-09-24
repo_root src/wyckoff_transformer/paths.py ran Apache-2.0 (permissive) · d04319a29ef33c57 · report fixture: signature start:PosixPath/0/? · returned PosixPath · level recorded 2026-09-24
service_ops_vectors src/wyckoff_transformer/preprocess_wychoffs.py ran fingerprinted Apache-2.0 (permissive) · eb8846d66ef5069d · report fixture: signature width:int/0/int · returned dict · level recorded 2026-09-24
start_log_prior src/wyckoff_transformer/gene_likelihood.py ran Apache-2.0 (permissive) · 4c607b6dfe2abca6 · report fixture: signature trainer:MockTrainer/0/?, records:DataFrame/0/? · returned Series · level recorded 2026-09-24
start_to_db_key src/wyckoff_transformer/csp.py ran fingerprinted Apache-2.0 (permissive) · dc734da5f9a66136 · report fixture: signature start:Tensor/1/float · returned float · level recorded 2026-09-24
warmup_stable_decay src/wyckoff_transformer/schedules.py ran Apache-2.0 (permissive) · a0fd18cdf3bb9181 · report fixture: signature optimizer:SGD/0/?, total_steps:int/0/int, warmup_fraction:float/0/float, decay_fraction:float/0/float, final_lr_fraction:float/0/float, decay_shape:str/0/str · returned LambdaLR · level recorded 2026-09-24
build_clean_relaxation_condition src/wyckoff_transformer/gene_energy.py unverified Apache-2.0 (permissive) · f2619508491644d0 · report
build_combinatorics src/wyckoff_transformer/csp.py unverified Apache-2.0 (permissive) · d8cfa8a7bd164a1f · report
build_tokenised_prediction_tensors src/wyckoff_transformer/prediction.py unverified Apache-2.0 (permissive) · 2efd8e80003f0994 · report
data_store src/wyckoff_transformer/paths.py unverified Apache-2.0 (permissive) · 6205bcf4c0b21416 · report
filter_supported_tokens src/wyckoff_transformer/prediction.py unverified Apache-2.0 (permissive) · 7682a09675335624 · report
tokeniser_key src/wyckoff_transformer/prediction.py unverified Apache-2.0 (permissive) · e28290d6a95905bf · report
gabeguo/cdvae_xrd community 1 sample, 1 ran (1 instrument-failed) · 0 unverified
build_mlp in 74 papers cdvae/pl_modules/model.py ran · our draft was wrong MIT (permissive) · 7a500e0992fceda5 · report fixture: signature in_dim:int/0/int, hidden_dim:int/0/int, fc_num_layers:int/0/int, out_dim:int/0/int · returned Sequential · level recorded, date not recorded
trachote/dp-cdvae community 1 sample, 1 ran (1 instrument-failed) · 0 unverified
optimization evaluate.py ran · our draft was wrong MIT (permissive) · 6bfe2286fd2a1f92 · report fixture: signature model:callable/0/?, prop_idx:int/0/int, ld_kwargs:dict/0/?, data_loader:list/1/Tensor, num_starting_points:int/0/int, num_gradient_steps:int/0/int, lr:float/0/float, num_saved_crys:int/0/int · returned dict · level recorded, date not recorded
ixsluo/cond-cdvae community 3 samples, 2 ran (2 instrument-failed) · 1 unverified
replace_hydride_formula scripts/evaluate.py ran · our draft was wrong fingerprinted MIT (permissive) · ca13c46956b131eb · report fixture: signature string:str/0/str · returned str · level recorded, date not recorded
sample_range scripts/evaluate.py ran · our draft was wrong fingerprinted MIT (permissive) · 51f77bb8655afae1 · report fixture: signature string:str/0/str · returned str · level recorded, date not recorded
get_file_paths in 14 papers scripts/compute_metrics.py unverified MIT (permissive) · 9a6539ae57f02d7a · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. For a generated assembly, an independent peer is a second implementation generated separately from the same paper text; neither implementation read the authors' code. At level 2 the two agreed numerically on the fixture's inputs; level 3 is recorded when property tests derived from the paper also passed on both, and each level 3 card prints how many its report lists. A generated assembly's label was set when its code was generated for the method, and no model judged it; its card prints it as "by construction" or "not judged by a model". Where an assembly's phrase has no constructed figure, a brick that ran is not among the harvested samples this build reads, and whether a brick's run only constructed an object is read from that brick's harvested sample, so the figure was not counted; it does not say there were none.

1 assembly on this paper: 0 harvested, 1 generated.

generated Assembly for Crystal Diffusion Variational Autoencoder (CDVAE) (graph method: Crystal Diffusion Variational Autoencoder (CDVAE)), generated from the paper text, not the repository's code. 1 brick; individually on a fixture: 1 ran (1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. Ran end-to-end against an independent peer (level 3; 5 property tests passed on both).

label matches by construction: the code was generated for this method (not judged by a model)

  1. langevin_coordinate_update entry · level 3 · ran · honoured contract · a895de13bb8e54d3aa9d9eab62c3e6353190585f4858bca60f565ed297752e04

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2110.06197") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

DecoderInductive BiasTranslation

1 archive task tag without a task page not shown.

Datasets

Introduced by this paper, per the archive.

MP20

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Diffusion

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections