SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1611.01734 · 2016

Deep Biaffine Attention for Neural Dependency Parsing

arXiv · PDF · Open in the Atlas

Code that ran

14 of 19 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; 1 honoured the contract we drafted, 1 violated it and 10 have no contract verdict; 2 failures of our instrument; 5 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/1611.01734.json · badge: /api/ran/1611.01734.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
chantera/biaffineparser community (classifier label)
classifier basis not recorded on the link
7 ran · 2 no recorded run
tdozat/Parser community (classifier label)
classifier basis not recorded on the link
2 ran · 1 no recorded run
baidu/DDParser community (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
redrew/neuralrst-bottom-up reimplementation (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
JoesSattes/Thai-Biaffine-Dependency-Parsing reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
wanghm92/Sing_Par community (classifier label)
classifier basis not recorded on the link
0 ran · 2 no recorded run
FunctionStatusWhere it lives
add_docstring Ran
ran on a synthesized input
baidu/DDParser/ddparser/ernie/file_utils.py
code served (permissive licence) · get_code("35a97b86a05fcd5c")
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 function. 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

def add_docstring(doc):
    def func(f):
        f.__doc__ += '\n======other docs from supper class ======\n%s' % doc
        return f
    return func

# Create a simple function with an initial docstring
def sample_function():
    """Initial documentation."""
    pass

# Define the arguments needed for add_docstring
doc_string = "This is additional documentation from the super class."

ARGS = {
    'doc': doc_string
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
append_name Ran
ran on a synthesized input
baidu/DDParser/ddparser/ernie/modeling_ernie.py
code served (permissive licence) · get_code("57d3f2c264025939")
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 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.

import torch

torch.manual_seed(0)

ARGS = {
    'name': 'test_model',
    'postfix': 'v1'
}
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 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
chuliu_edmonds Ran
ran on a synthesized input
chantera/biaffineparser/src/utils/chuliu_edmonds.py
code served (permissive licence) · get_code("906bfaff2c598764")
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 ndarray, ndim 1, dtype family int, shape (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 numpy as np

# Create a small scores matrix for the chuliu_edmonds function
# The function expects a 2D numpy array of scores
# Shape should be (n_nodes, n_nodes) where n_nodes is at least 2

np.random.seed(0)

# Create a simple scores matrix with shape (4, 4)
# This represents scores for edges in a directed graph
scores = np.random.randn(4, 4).astype(np.float32)

# Ensure it's a valid input by making it numeric and not all identical
scores = scores * 10  # Scale for better numerical behavior

ARGS = {
    "scores": scores
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family int, shape (4,) — shape (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
chuliu_edmonds_one_root Ran
ran on a synthesized input
chantera/biaffineparser/src/utils/chuliu_edmonds.py
code served (permissive licence) · get_code("f209dc6a414e163b")
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 ndarray, ndim 1, dtype family int, shape (5,) 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 numpy as np

# Create a small scores matrix for testing chuliu_edmonds_one_root
# The function expects a 2D array of shape (n, n) representing edge scores
# in a directed graph where we want to find the maximum spanning arborescence

# Create a simple 5x5 scores matrix with valid edge scores
scores = np.array([
    [0.0, 1.0, 2.0, 3.0, 4.0],
    [5.0, 0.0, 1.5, 2.5, 3.5],
    [4.0, 5.0, 0.0, 1.0, 2.0],
    [3.0, 4.0, 5.0, 0.0, 1.5],
    [2.0, 3.0, 4.0, 5.0, 0.0]
], dtype=np.float64)

ARGS = {
    "scores": scores
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family int, shape (5,) — shape (5,) 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
collate Ran
ran on a synthesized input
chantera/biaffineparser/src/data.py
code served (permissive licence) · get_code("fff6c1cc919ef43f")
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 list. 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

torch.manual_seed(0)

# The function collate takes:
# 1. batch: Iterable[Sequence[torch.Tensor]] - an iterable of sequences of tensors
# 2. device: Optional[torch.device] - an optional device to move tensors to

# Create a small batch of data
# Each row is a sequence of tensors (e.g., multiple fields per sample)
batch = [
    [torch.randn(2, 4), torch.randn(3, 4)],  # First sample: 2 tensors
    [torch.randn(2, 4), torch.randn(3, 4)],  # Second sample: 2 tensors
]

# Device is optional, test with None (CPU)
device = None

ARGS = {
    "batch": batch,
    "device": device,
}
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 list — a list 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
forward Ran
ran on a synthesized input
chantera/biaffineparser/src/training.py
code served (permissive licence) · get_code("2a8f48dd9ad2f576")
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 dict. 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 DummyModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.training = True
        self.linear_head = nn.Linear(4, 5)
        self.linear_deprel = nn.Linear(4, 6)
    
    def forward(self, x1, x2):
        # x1, x2 are input tensors
        logits_head = self.linear_head(x1)
        logits_deprel = self.linear_deprel(x2)
        lengths = torch.tensor([2, 2], dtype=torch.long)
        return logits_head, logits_deprel, lengths
    
    def compute_metrics(self, logits_head, logits_deprel, heads, deprels):
        # Return a dict with loss metrics
        return {
            'head_loss': torch.tensor(0.5),
            'deprel_loss': torch.tensor(0.3),
            'head_accuracy': torch.tensor(0.8),
            'deprel_accuracy': torch.tensor(0.75)
        }
    
    def decode(self, logits_head, logits_deprel, lengths):
        # Return predicted heads and deprels
        batch_size = logits_head.shape[0]
        predicted_heads = torch.randint(0, 5, (batch_size, 3))
        predicted_deprels = torch.randint(0, 6, (batch_size, 3))
        return predicted_heads, predicted_deprels

# Create model instance
model = DummyModel()

# Create batch: (*xs, heads, deprels)
# xs will be unpacked as x1, x2
x1 = torch.randn(2, 4)
x2 = torch.randn(2, 4)
heads = torch.tensor([[1, 2, 0], [0, 1, 2]], dtype=torch.long)
deprels = torch.tensor([[1, 2, 0], [0, 1, 2]], dtype=torch.long)

batch = (x1, x2, heads, deprels)

ARGS = {
    'model': model,
    'batch': batch
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
get_num_other_elem Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
redrew/neuralrst-bottom-up/models/architecture/bottom_up_architecture.py
pointer only (licence: NONE) · get_code("73d6747bfa5543ce")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, redrew/neuralrst-bottom-up 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.

is_number Ran
ran, violated the contract we drafted
JoesSattes/Thai-Biaffine-Dependency-Parsing/preprocess.py
code served (permissive licence) · get_code("1889abe3c29f3531")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned bool, 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.

ARGS = {
    's': '123.45'
}
Setup block drafted by
a language model, Claude Sonnet 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 bool, 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
load_embeddings Ran
ran on a synthesized input
chantera/biaffineparser/src/utils/data.py
code served (permissive licence) · get_code("d9f663d6651746c4")
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 OrderedDict. 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 tempfile
import os

# Create temporary files for testing
temp_dir = tempfile.mkdtemp()

# Create an embeddings file with format: token vector1 vector2 vector3
embeddings_file = os.path.join(temp_dir, "embeddings.txt")
with open(embeddings_file, 'w') as f:
    f.write("hello 0.1 0.2 0.3\n")
    f.write("world 0.4 0.5 0.6\n")

# Create a vocab file with one token per line
vocab_file = os.path.join(temp_dir, "vocab.txt")
with open(vocab_file, 'w') as f:
    f.write("token1\n")
    f.write("token2\n")

# Create separate embeddings file for vocab_file mode
embeddings_file2 = os.path.join(temp_dir, "embeddings2.txt")
with open(embeddings_file2, 'w') as f:
    f.write("0.1 0.2 0.3\n")
    f.write("0.4 0.5 0.6\n")

ARGS = {
    "file": embeddings_file,
    "vocab_file": None,
    "delimiter": " "
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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 OrderedDict — a OrderedDict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a OrderedDict 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
parse_conll Ran
ran on a synthesized input
chantera/biaffineparser/src/utils/conll.py
code served (permissive licence) · get_code("d9eca1a4508f789e")
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 generator. 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.

# Analyzing parse_conll function:
# - Takes 'lines' parameter which is iterated over
# - Each line is stripped and checked for various conditions
# - Empty lines trigger a yield
# - Lines starting with '#' are skipped
# - Other lines are split by tabs and parsed into token dictionaries

# The function expects CoNLL format input with tab-separated columns
# Standard CoNLL-U format has 10 columns, but this code uses 8:
# id, form, lemma, cpostag, postag, feats, head, deprel

# Creating sample CoNLL formatted lines
lines = [
    "# sent_id = 1",
    "# text = The cat sat.",
    "1\tThe\tthe\tDET\tDT\t_\t2\tdet",
    "2\tcat\tcat\tNOUN\tNN\t_\t3\tnsubj",
    "3\tsat\tsit\tVERB\tVBD\t_\t0\troot",
    "4\t.\t.\tPUNCT\t.\t_\t3\tpunct",
    "",
    "# sent_id = 2",
    "# text = Dogs run.",
    "1\tDogs\tdog\tNOUN\tNNS\t_\t2\tnsubj",
    "2\trun\trun\tVERB\tVBP\t_\t0\troot",
    "3\t.\t.\tPUNCT\t.\t_\t2\tpunct",
    "",
]

ARGS = {
    'lines': lines
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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 generator — a generator output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a generator 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
set_label_action Ran
ran; our contract draft was wrong, not the code
redrew/neuralrst-bottom-up/train_rst_parser.py
pointer only (licence: NONE) · get_code("06e98e0817b9c50f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, redrew/neuralrst-bottom-up 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.

sigmoid Ran
ran on a synthesized input
tdozat/Parser/lib/linalg.py
code served (permissive licence) · get_code("5ee5f3de417ac23d")
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 EagerTensor, 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 tensorflow as tf
import numpy as np

def sigmoid(x):
    return (tf.tanh(x) + 1) / 2

# Create a small TensorFlow tensor as input
x_input = tf.constant([[1.0, -0.5, 2.0, -1.5],
                       [0.5, -1.0, 1.5, -0.5]], dtype=tf.float32)

ARGS = {
    "x": x_input
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, 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
tanh Ran
ran, honoured the contract we drafted
tdozat/Parser/lib/linalg.py
code served (permissive licence) · get_code("00edc2dc619cd019")
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 EagerTensor, 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 tensorflow as tf
import numpy as np

# Create a small TensorFlow tensor for the input
x = tf.constant([[1.0, -0.5, 0.2, 1.5], [0.8, -1.2, 0.3, -0.7]], dtype=tf.float32)

# Construct the arguments dict
ARGS = {
    'x': x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, 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
tarjan Ran
ran on a synthesized input
chantera/biaffineparser/src/utils/chuliu_edmonds.py
code served (permissive licence) · get_code("02240c2030c902b7")
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 list, ndim 1, 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 numpy as np

# The function tarjan takes a tree parameter which is a numpy array
# It's used with np.where, np.equal, indexing with integers
# The array represents parent indices or dependencies

# Create a small tree structure as a numpy array
# tree[i] represents the parent/dependent of node i
# Example: tree = [0, 0, 1, 2] means:
#   node 0 -> 0 (root)
#   node 1 -> 0 (parent is 0)
#   node 2 -> 1 (parent is 1)
#   node 3 -> 2 (parent is 2)

tree = np.array([0, 0, 1, 2], dtype=np.int64)

ARGS = {
    "tree": tree
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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
categorical_accuracy Not yet run
unverified (harvested, no recorded run)
chantera/biaffineparser/src/models.py
code served (permissive licence) · get_code("4cb6893c39bb6390")
How it ran

no invocation recorded.

evaluate Not yet run
unverified (harvested, no recorded run)
chantera/biaffineparser/src/utils/conll.py
code served (permissive licence) · get_code("75843fe2dd997b53")
How it ran

no invocation recorded.

orthonormal_initializer Not yet run
unverified (harvested, no recorded run)
tdozat/Parser/lib/linalg.py
code served (permissive licence) · get_code("2e71814e529d090b")
How it ran

no invocation recorded.

orthonormal_initializer Not yet run
unverified (harvested, no recorded run)
wanghm92/Sing_Par/lib/linalg.py
code served (permissive licence) · get_code("90d46b3cd3eacbaf")
How it ran

no invocation recorded.

smooth Not yet run
unverified (harvested, no recorded run)
wanghm92/Sing_Par/lib/etc/smooth.py
code served (permissive licence) · get_code("0073d56ba1481305")
How it ran

no invocation recorded.

Also present in other papers' code

Of 19 samples lifted from this paper's linked repositories, 19 carry a function identity and 0 do not. Of the 19 distinct functions with an identity, 3 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 3 later) and 16 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
tanh 2 other papers
2 later
1803.06585 (2018-03)
Learning Long Term Dependencies via Fourier Recurrent Units
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
orthonormal_initializer 1 other paper
1 later
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
sigmoid 1 other paper
1 later
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

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

This paper builds off recent work from Kiperwasser & Goldberg (2016) using neural attention in a simple graph-based dependency parser. We use a larger but more thoroughly regularized parser than other recent BiLSTM-based approaches, with biaffine classifiers to predict arcs and labels. Our parser gets state of the art or near state of the art performance on standard treebanks for six different languages, achieving 95.7% UAS and 94.1% LAS on the most popular English PTB dataset. This makes it the highest-performing graph-based parser on this benchmark---outperforming Kiperwasser Goldberg (2016) by 1.8% and 2.2%---and comparable to the highest performing transition-based parser (Kuncoro et al., 2016), which achieves 95.8% UAS and 94.6% LAS. We also show which hyperparameter choices had a significant effect on parsing accuracy, allowing us to achieve large gains over other graph-based approaches.

For agents

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

get_harvested_code_for_paper("1611.01734")
get_code_for_paper("1611.01734")
have("1611.01734")

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

curl https://syntology.ai/api/ran/1611.01734.json

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

[![Syntology run record](https://syntology.ai/api/ran/1611.01734.svg)](https://syntology.ai/paper/1611.01734)

Connect an agent — have() is free.