SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2605.07048 · 2026

Unlocking High-Fidelity Molecular Generation from Mass Spectra via Dual-Stream Line Graph Diffusion

Depeng Xu, Xujun Che

arXiv · PDF

Code that ran

11 of 16 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (6 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 0 violated it and 9 have no contract verdict; 0 failures of our instrument; 5 no recorded run), measured 2026-11-22; 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/2605.07048.json · badge: /api/ran/2605.07048.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
du-lab-data-science/DualLGD community
Papers with Code archive
11 ran · 5 no recorded run
FunctionStatusWhere it lives
ELUPlusOneSelfAttention Ran
ran on a synthesized input
du-lab-data-science/DualLGD/src/models/duallgd_transformer.py
code served (permissive licence) · get_code("7631aefea4a1a618")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 3, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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

# INIT_ARGS for ELUPlusOneSelfAttention.__init__
INIT_ARGS = {
    'dim': 8,
    'num_heads': 2,
    'dropout': 0.1
}

# ARGS for ELUPlusOneSelfAttention.forward
ARGS = {
    'x': torch.randn(2, 5, 8),  # (batch_size=2, num_tokens=5, dim=8)
    'token_mask': torch.ones(2, 5)  # (batch_size=2, num_tokens=5)
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 3, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
FeedForward Ran
ran on a synthesized input · constructed an object; no result was computed
du-lab-data-science/DualLGD/src/models/duallgd_transformer.py
code served (permissive licence) · get_code("6ce3de746b63cb89")
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 an instance of its own class, FeedForward: the run constructed the object and did not call it, so no result was computed. 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.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "dim": 4,
    "hidden_dim": 8,
    "dropout": 0.1,
}
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 FeedForward — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a FeedForward 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
FiLMConditioner Ran
ran on a synthesized input · constructed an object; no result was computed
du-lab-data-science/DualLGD/src/models/duallgd_transformer.py
code served (permissive licence) · get_code("e3aad312f5e73601")
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 an instance of its own class, FiLMConditioner: the run constructed the object and did not call it, so no result was computed. 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.

import torch
import torch.nn as nn

torch.manual_seed(0)

class FiLMConditioner(nn.Module):

    def __init__(self, feature_dim: int, cond_dim: int):
        super().__init__()
        self.gamma = nn.Linear(cond_dim, feature_dim)
        self.beta = nn.Linear(cond_dim, feature_dim)

    def forward(self, x: torch.Tensor, cond: torch.Tensor) -> torch.Tensor:
        gamma = self.gamma(cond).unsqueeze(1)
        beta = self.beta(cond).unsqueeze(1)
        return (1.0 + gamma) * x + beta

# Create a small instance of FiLMConditioner
feature_dim = 4
cond_dim = 3
model = FiLMConditioner(feature_dim=feature_dim, cond_dim=cond_dim)

# Create small test tensors
# x: (batch_size=2, sequence_length=1, feature_dim=4)
x = torch.randn(2, 1, 4)
# cond: (batch_size=2, cond_dim=3)
cond = torch.randn(2, 3)

ARGS = {
    "feature_dim": feature_dim,
    "cond_dim": cond_dim
}
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 FiLMConditioner — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a FiLMConditioner 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
MaskedCrossAttention Ran
ran on a synthesized input
du-lab-data-science/DualLGD/src/models/duallgd_transformer.py
code served (permissive licence) · get_code("bffc610dbe3fc42d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 3, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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

# INIT_ARGS for MaskedCrossAttention.__init__
INIT_ARGS = {
    'q_dim': 8,
    'kv_dim': 6,
    'num_heads': 2,
    'dropout': 0.1
}

# ARGS for MaskedCrossAttention.forward
batch_size = 2
q_tokens = 5
kv_tokens = 7
q_dim = INIT_ARGS['q_dim']
kv_dim = INIT_ARGS['kv_dim']

q_x = torch.randn(batch_size, q_tokens, q_dim)
kv_x = torch.randn(batch_size, kv_tokens, kv_dim)

# q_mask: boolean mask for query tokens (batch_size, q_tokens)
q_mask = torch.ones(batch_size, q_tokens, dtype=torch.bool)
q_mask[0, 3:] = False  # mask out last 2 tokens in first batch
q_mask[1, 4:] = False  # mask out last token in second batch

# kv_mask: boolean mask for key-value tokens (batch_size, kv_tokens)
kv_mask = torch.ones(batch_size, kv_tokens, dtype=torch.bool)
kv_mask[0, 5:] = False  # mask out last 2 tokens in first batch
kv_mask[1, 6:] = False  # mask out last token in second batch

# local_mask: pairwise attention mask (batch_size, q_tokens, kv_tokens)
local_mask = torch.ones(batch_size, q_tokens, kv_tokens, dtype=torch.bool)
# Create some local attention pattern (e.g., each query can only attend to nearby keys)
for b in range(batch_size):
    for i in range(q_tokens):
        for j in range(kv_tokens):
            if abs(i - j) > 3:  # only attend within window of 3
                local_mask[b, i, j] = False

ARGS = {
    'q_x': q_x,
    'kv_x': kv_x,
    'q_mask': q_mask,
    'kv_mask': kv_mask,
    'local_mask': local_mask,
    'return_attn': False
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 3, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
NodeEdgeBlock Ran
ran on a synthesized input · constructed an object; no result was computed
du-lab-data-science/DualLGD/src/models/duallgd_transformer.py
code served (permissive licence) · get_code("455e20737ca978e1")
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 an instance of its own class, NodeEdgeBlock: the run constructed the object and did not call it, so no result was computed. 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.

import torch
import torch.nn as nn
import math

torch.manual_seed(0)

# Create a minimal stand-in for diffusion_utils
class diffusion_utils:
    @staticmethod
    def assert_correctly_masked(tensor, mask):
        pass

# Inject it into the module namespace so forward() can call it
import sys
sys.modules['diffusion_utils'] = diffusion_utils

# Define dimensions (distinct sizes for different purposes)
dx = 8      # node feature dimension
de = 6      # edge feature dimension
dy = 4      # global feature dimension
n_head = 2  # number of attention heads
bs = 2      # batch size
n = 3       # number of nodes

# Create tensors with appropriate shapes and values
X = torch.randn(bs, n, dx)
E = torch.randn(bs, n, n, de)
y = torch.randn(bs, dy)
node_mask = torch.ones(bs, n, dtype=torch.bool)

ARGS = {
    'dx': dx,
    'de': de,
    'dy': dy,
    'n_head': n_head,
}
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 NodeEdgeBlock — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a NodeEdgeBlock 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
PerformerFeatureMap Ran
ran on a synthesized input · constructed an object; no result was computed
du-lab-data-science/DualLGD/src/models/duallgd_transformer.py
code served (permissive licence) · get_code("655c597bf5fe6021")
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 an instance of its own class, PerformerFeatureMap: the run constructed the object and did not call it, so no result was computed. 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.

import torch
import torch.nn as nn
import math

torch.manual_seed(0)

ARGS = {
    'head_dim': 4,
    'num_features': 8,
    'kernel': 'positive',
    'feature_type': 'orthogonal',
    'redraw_on_forward': False
}
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 PerformerFeatureMap — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a PerformerFeatureMap 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
PerformerSelfAttention Ran
ran on a synthesized input · constructed an object; no result was computed
du-lab-data-science/DualLGD/src/models/duallgd_transformer.py
code served (permissive licence) · get_code("2bfc8f7589e18ce3")
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 an instance of its own class, PerformerSelfAttention: the run constructed the object and did not call it, so no result was computed. 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.

import torch
import torch.nn as nn
import math

torch.manual_seed(0)

ARGS = {
    'dim': 8,
    'num_heads': 2,
    'dropout': 0.1,
    'performer_num_features': 4,
    'performer_kernel': 'positive',
    'performer_feature_type': 'orthogonal',
    'performer_redraw_on_forward': False,
}
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 PerformerSelfAttention — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a PerformerSelfAttention 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
SoftmaxSelfAttention Ran
ran on a synthesized input
du-lab-data-science/DualLGD/src/models/duallgd_transformer.py
code served (permissive licence) · get_code("756b84fee8389e11")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 3, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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

# INIT_ARGS for SoftmaxSelfAttention.__init__
INIT_ARGS = {
    'dim': 8,
    'num_heads': 2,
    'dropout': 0.1
}

# ARGS for SoftmaxSelfAttention.forward
ARGS = {
    'x': torch.randn(2, 5, 8),  # (batch_size=2, num_tokens=5, dim=8)
    'token_mask': torch.tensor([
        [True, True, True, False, False],
        [True, True, True, True, False]
    ])  # (batch_size=2, num_tokens=5)
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 3, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
Xtoy Ran
ran on a synthesized input · constructed an object; no result was computed
du-lab-data-science/DualLGD/src/models/duallgd_transformer.py
code served (permissive licence) · get_code("267ec92b6214cdcb")
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 an instance of its own class, Xtoy: the run constructed the object and did not call it, so no result was computed. 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.

import torch
import torch.nn as nn

torch.manual_seed(0)

class Xtoy(nn.Module):

    def __init__(self, dx, dy):
        """ Map node features to global features """
        super().__init__()
        self.lin = nn.Linear(4 * dx, dy)

    def forward(self, X, x_mask):
        """ X: bs, n, dx. """
        x_mask = x_mask.expand(-1, -1, X.shape[-1])
        float_imask = 1 - x_mask.float()
        m = X.sum(dim=1) / torch.sum(x_mask, dim=1)
        mi = (X + 100000.0 * float_imask).min(dim=1)[0]
        ma = (X - 100000.0 * float_imask).max(dim=1)[0]
        std = torch.sum((X - m[:, None, :]) ** 2 * x_mask, dim=1) / torch.sum(x_mask, dim=1)
        z = torch.hstack((m, mi, ma, std))
        out = self.lin(z)
        return out

# Create test data
batch_size = 2
num_nodes = 3
dx = 4
dy = 5

# X: bs, n, dx
X = torch.randn(batch_size, num_nodes, dx)

# x_mask: bs, n, 1 (will be expanded to bs, n, dx in forward)
x_mask = torch.ones(batch_size, num_nodes, 1, dtype=torch.bool)

ARGS = {
    "dx": dx,
    "dy": dy,
}
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 Xtoy — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Xtoy 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
_iid_gaussian_random_matrix Ran
ran, honoured the contract we drafted
du-lab-data-science/DualLGD/src/models/duallgd_transformer.py
code served (permissive licence) · get_code("59d4282bb5cdc95d")
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 2, dtype family float, shape (2, 4) 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)

ARGS = {
    "num_features": 2,
    "head_dim": 4,
    "device": torch.device("cpu"),
    "dtype": torch.float32,
}
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 2, dtype family float, shape (2, 4) — shape (2, 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
_orthogonal_gaussian_random_matrix Ran
ran, honoured the contract we drafted
du-lab-data-science/DualLGD/src/models/duallgd_transformer.py
code served (permissive licence) · get_code("93217a07da57bba8")
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 2, dtype family float, shape (8, 4) 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 math

torch.manual_seed(0)

ARGS = {
    'num_features': 8,
    'head_dim': 4,
    'device': torch.device('cpu'),
    'dtype': torch.float32,
}
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 2, dtype family float, shape (8, 4) — shape (8, 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
DualAttentionBlock Not yet run
unverified (harvested, no recorded run)
du-lab-data-science/DualLGD/src/models/duallgd_transformer.py
code served (permissive licence) · get_code("795cd5cd9061fb80")
How it ran

no invocation recorded.

DualLGD Not yet run
unverified (harvested, no recorded run)
du-lab-data-science/DualLGD/src/models/duallgd_transformer.py
code served (permissive licence) · get_code("f442cb7661864476")
How it ran

no invocation recorded.

Etoy Not yet run
unverified (harvested, no recorded run)
du-lab-data-science/DualLGD/src/models/duallgd_transformer.py
code served (permissive licence) · get_code("11d64ebd668b37b1")
How it ran

no invocation recorded.

XEyTransformerLayer Not yet run
unverified (harvested, no recorded run)
du-lab-data-science/DualLGD/src/models/duallgd_transformer.py
code served (permissive licence) · get_code("58397444f9e72fe0")
How it ran

no invocation recorded.

masked_softmax Not yet run
unverified (harvested, no recorded run)
du-lab-data-science/DualLGD/src/models/duallgd_transformer.py
code served (permissive licence) · get_code("bab87e7f0fea40be")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 0 later) and 15 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
XEyTransformerLayer
brick of this paper's assembly: helper
2 other papers
2 dated earlier
2209.14734 (2022-09)
DiGress: Discrete Denoising diffusion for graph generation
2403.16883 (2024-03)
GLAD: Improving Latent Graph Generative Modeling with Simple Quantization

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

De novo molecular generation from tandem mass spectra is a challenging inverse problem whose core difficulty lies in the circular dependency between atom-level and bond-level reasoning: determining a bond's type requires knowing its endpoint atoms' chemical environment, yet an atom's environment is in turn defined by its incident bonds. Existing graph diffusion methods process atoms and bonds within a single computation stream, where atom-bond information synchronization can only occur implicitly across layers. We argue that this single-stream paradigm, rather than the choice of any particular aggregation kernel, is a key architectural bottleneck. We propose DualLGD (Dual-stream Line Graph Diffusion), which reformulates molecular graph denoising as the alternating solution of two coupled subproblems: atom-level reasoning and bond-level reasoning, each operating in its own dedicated representation space. The line graph provides a natural mathematical construction for the bond space, in which bond angles, dihedrals, conjugation chains, and rings correspond to local topological motifs between bonds. Incidence-constrained bidirectional cross-attention synchronizes the two streams at every layer, ensuring that each atom attends only to its incident bonds and vice versa, respecting the fundamental chemical principle that an atom's environment is determined by its bonding context. On the NPLIB1 and MassSpecGym benchmarks, DualLGD achieves top-1 accuracy of 34.37% and 23.89%, approximately 3× the previous state of the art. Ablation studies confirm the architecture as the primary source of improvement: DualLGD without any pre-training already surpasses the previous best fully pretrained model. Code and pretrained models are publicly available at https://github.com/du-lab-data-science/DualLGD.

For agents

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

get_harvested_code_for_paper("2605.07048")
get_code_for_paper("2605.07048")
have("2605.07048")

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

curl https://syntology.ai/api/ran/2605.07048.json

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

[![Syntology run record](https://syntology.ai/api/ran/2605.07048.svg)](https://syntology.ai/paper/2605.07048)

Connect an agent — have() is free.