SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.17005 · NeurIPS · 2024

Hybrid Generative AI for De Novo Design of Co-Crystals with Enhanced Tabletability

Nina Gubina, Andrei Dmitrenko, Gleb Solovev, Lyubov Yamshchikova, Oleg Petrov, Ivan Lebedev, Nikita Serov, Grigorii Kirgizov, Nikolay Nikitin, Vladimir Vinogradov

arXiv · PDF · Open in the Atlas

Code that ran

7 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 4 failures of our instrument; 2 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2410.17005.json · badge: /api/ran/2410.17005.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
ai-chem/GEMCODE canonical (classifier label)
classifier basis: census
5 ran · 1 no recorded run
ai-chem/gemcode canonical (classifier label)
classifier basis: classifier_2025
2 ran · 1 no recorded run
FunctionStatusWhere it lives
KLAnnealer Ran
ran, honoured the contract we drafted
ai-chem/gemcode/generative_models/TVAE/train_VAE/train_phys.py
code served (permissive licence) · get_code("daec152adb204c8a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

class OptConfig:
    def __init__(self):
        self.KLA_ini_beta = 0.0
        self.KLA_inc_beta = 0.1
        self.KLA_beg_epoch = 5

opt = OptConfig()
epoch = 10

ARGS = {
    "opt": opt,
    "epoch": epoch
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
attention Ran
ran; our contract draft was wrong, not the code
ai-chem/GEMCODE/generative_models/TVAE/Sublayers.py
code served (permissive licence) · get_code("e7d0985e55efb9e7")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 5, dtype family float, shape (2, 1, 2, 5, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import math

torch.manual_seed(0)

# Analyze the function signature and usage:
# - q, k, v: tensors used in matmul operations with transpose
# - d_k: scalar (int) used in sqrt for scaling
# - mask: optional tensor used with unsqueeze(1) and masked_fill
# - dropout: optional callable (nn.Module) applied to scores

# Create small distinct-sized tensors for q, k, v
# q shape: (batch=2, seq_len=5, d_k=4)
# k shape: (batch=2, seq_len=6, d_k=4)
# v shape: (batch=2, seq_len=6, d_v=3)
batch_size = 2
seq_len_q = 5
seq_len_k = 6
d_k = 4
d_v = 3

q = torch.randn(batch_size, seq_len_q, d_k)
k = torch.randn(batch_size, seq_len_k, d_k)
v = torch.randn(batch_size, seq_len_k, d_v)

# mask: shape (batch=2, 1, seq_len_q=5, seq_len_k=6)
# Values of 1 indicate valid positions, 0 indicates masked positions
mask = torch.ones(batch_size, 1, seq_len_q, seq_len_k)
# Mask out last position for demonstration
mask[:, :, :, -1] = 0

# dropout: an nn.Dropout module (has __call__ method)
dropout = nn.Dropout(p=0.1)

ARGS = {
    "q": q,
    "k": k,
    "v": v,
    "d_k": d_k,
    "mask": mask,
    "dropout": dropout,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 5, dtype family float, shape (2, 1, 2, 5, 3) — shape (2, 1, 2, 5, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
batch_size_fn Ran
ran, honoured the contract we drafted
ai-chem/GEMCODE/generative_models/TVAE/Batch.py
code served (permissive licence) · get_code("71da2b1966ba54e3")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned int, ndim 0, dtype family int, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

class BatchItem:
    def __init__(self, src, trg):
        self.src = src
        self.trg = trg

new = BatchItem(src=[1, 2, 3], trg=[4, 5])
count = 1
sofar = 0

ARGS = {
    "new": new,
    "count": count,
    "sofar": sofar
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, ndim 0, dtype family int, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
create_masks Ran
ran; our contract draft was wrong, not the code
ai-chem/GEMCODE/generative_models/TVAE/Batch.py
code served (permissive licence) · get_code("978ebdafdd719af8")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
from torch.autograd import Variable

torch.manual_seed(0)

class MockOpt:
    def __init__(self):
        self.src_pad = 0
        self.trg_pad = 0
        self.cond_dim = 3
        self.use_cond2dec = True
        self.use_cond2lat = True
        self.device = 'cpu'

opt = MockOpt()

src = torch.randint(1, 5, (2, 4))
trg = torch.randint(1, 5, (2, 3))
cond = torch.randint(1, 5, (2, 3))

ARGS = {
    'src': src,
    'trg': trg,
    'cond': cond,
    'opt': opt
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_clones Ran
ran; our contract draft was wrong, not the code
ai-chem/GEMCODE/generative_models/TVAE/Models.py
pointer only (licence: MIT); the stored copy get_code serves, from arshadshk/SAINT-pytorch, is pointer only (licence: NONE) · get_code("891b8ebab395921f")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from arshadshk/SAINT-pytorch, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

nopeak_mask Ran
ran, violated the contract we drafted
ai-chem/GEMCODE/generative_models/TVAE/Batch.py
code served (permissive licence) · get_code("592078669d81c3ad")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 3, dtype family float, shape (1, 4, 4) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
from torch.autograd import Variable

torch.manual_seed(0)

class OptConfig:
    def __init__(self, use_cond2dec=False, cond_dim=3, device='cpu'):
        self.use_cond2dec = use_cond2dec
        self.cond_dim = cond_dim
        self.device = device

# Test case 1: without conditional to decoder
opt_simple = OptConfig(use_cond2dec=False, cond_dim=3, device='cpu')

# Test case 2: with conditional to decoder
opt_cond = OptConfig(use_cond2dec=True, cond_dim=3, device='cpu')

ARGS = {
    'size': 4,
    'opt': opt_simple
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (1, 4, 4) — shape (1, 4, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
yesno Ran
ran; our contract draft was wrong, not the code
ai-chem/gemcode/generative_models/TVAE/train_VAE/train_phys.py
code served (permissive licence) · get_code("9cf9fdc863be17f3")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

# The function yesno takes a response parameter and validates it
# It expects a string that is either 'y' or 'n'
# For setup, we provide a valid response string

ARGS = {
    'response': 'y'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
k_best_outputs Not yet run
unverified (harvested, no recorded run)
ai-chem/GEMCODE/generative_models/TVAE/Beam.py
code served (permissive licence) · get_code("46805a58d000dc41")
How it ran

no invocation recorded.

loss_function Not yet run
unverified (harvested, no recorded run)
ai-chem/gemcode/generative_models/TVAE/train_VAE/train_phys.py
code served (permissive licence) · get_code("b0f7d8326b02cb76")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 1 the same month, 1 later) and 7 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
get_clones 40 other papers
32 dated earlier, 1 the same month, 6 later, 1 same year, order not determinable · 8 only as an assembly brick
1905.12794 (2019-05)
Fashion IQ: A New Dataset Towards Retrieving Images by Natural Language Feedback
2002.07033 (2020-02)
Towards an Appropriate Query, Key, and Value Computation for Knowledge Tracing
2010.12042 (2020-10)
SAINT+: Integrating Temporal Features for EdNet Correctness Prediction
and 37 more not named here
attention 4 other papers
4 dated earlier
1905.12794 (2019-05)
Fashion IQ: A New Dataset Towards Retrieving Images by Natural Language Feedback
2009.13603 (2020-09)
Visual Pivoting for (Unsupervised) Entity Alignment
2202.04298 (2022-02)
Image Difference Captioning with Pre-training and Contrastive Learning
and 1 more not named here

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Co-crystallization is an accessible way to control physicochemical characteristics of organic crystals, which finds many biomedical applications. In this work, we present Generative Method for Co-crystal Design (GEMCODE) 1 , a novel pipeline for automated co-crystal screening based on the hybridization of deep generative models and evolutionary optimization for broader exploration of the target chemical space. GEMCODE enables fast de novo co-crystal design with target tabletability profiles, which is crucial for the development of pharmaceuticals. With a series of experimental studies highlighting validation and discovery cases, we show that GEMCODE is effective even under realistic computational constraints. Furthermore, we explore the potential of language models in generating co-crystals. Finally, we present numerous previously unknown co-crystals predicted by GEM-CODE and discuss its potential in accelerating drug development.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2410.17005")
get_code_for_paper("2410.17005")
have("2410.17005")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2410.17005.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2410.17005.svg)](https://syntology.ai/paper/2410.17005)

Connect an agent — have() is free.