SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1902.10909 · 2019

BERT for Joint Intent Classification and Slot Filling

arXiv · PDF · Open in the Atlas

Code that ran

26 of 32 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, 0 violated it and 16 have no contract verdict; 9 failures of our instrument; 6 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/1902.10909.json · badge: /api/ran/1902.10909.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
alibaba-damo-academy/spokennlp community (classifier label)
classifier basis not recorded on the link
8 ran · 3 no recorded run
Polly42Rose/SiriusIntentPredictionSlotFilling community (classifier label)
classifier basis not recorded on the link
6 ran · 2 no recorded run
monologg/JointBERT community (classifier label)
classifier basis: plan_row
5 ran · 0 no recorded run
mangushev/intent_slot community (classifier label)
classifier basis not recorded on the link
4 ran · 1 no recorded run
yinghao1019/Joint_learn reimplementation (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
dsindex/iclassifier reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
abridge_model_name Ran
ran on a synthesized input
alibaba-damo-academy/spokennlp/emnlp2023-topic_segmentation/src/utils.py
code served (permissive licence) · get_code("0628b6fa4fa6d748")
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 os

def abridge_model_name(model_name_or_path):
    model_name = os.path.basename(model_name_or_path)
    if 'longformer' in model_name_or_path:
        model_name = 'lf'
    elif 'bigbird' in model_name_or_path:
        model_name = 'bb'
    elif 'bert' in model_name_or_path:
        model_name = 'bert'
    elif 'electra' in model_name_or_path:
        model_name = 'ele'
    else:
        raise ValueError('not supported model_name')
    return model_name

ARGS = {
    'model_name_or_path': 'longformer-base-4096'
}
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 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
convert_examples_to_features Ran
ran on a synthesized input
monologg/JointBERT/data_loader.py
code served (permissive licence) · get_code("9d27d5015d7a3b3e")
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 copy
import json
import logging

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

class MockExample:
    """Stand-in for an example object."""
    def __init__(self, guid, words, slot_labels, intent_label):
        self.guid = guid
        self.words = words
        self.slot_labels = slot_labels
        self.intent_label = intent_label

class MockTokenizer:
    """Stand-in for a tokenizer."""
    def __init__(self):
        self.cls_token = "[CLS]"
        self.sep_token = "[SEP]"
        self.unk_token = "[UNK]"
        self.pad_token_id = 0
        self._token_to_id = {
            "[CLS]": 101,
            "[SEP]": 102,
            "[UNK]": 100,
            "hello": 1,
            "world": 2,
            "i": 3,
            "am": 4,
            "test": 5,
        }
    
    def tokenize(self, word):
        """Tokenize a word into subword tokens."""
        word_lower = word.lower()
        if word_lower in self._token_to_id:
            return [word_lower]
        return []
    
    def convert_tokens_to_ids(self, tokens):
        """Convert tokens to their IDs."""
        return [self._token_to_id.get(token, 100) for token in tokens]

examples = [
    MockExample(
        guid="example-1",
        words=["hello", "world"],
        slot_labels=["0", "1"],
        intent_label="2"
    ),
    MockExample(
        guid="example-2",
        words=["i", "am", "test"],
        slot_labels=["1", "0", "1"],
        intent_label="1"
    ),
]

tokenizer = MockTokenizer()

ARGS = {
    "examples": examples,
    "max_seq_len": 16,
    "tokenizer": tokenizer,
    "pad_token_label_id": -100,
    "cls_token_segment_id": 0,
    "pad_token_segment_id": 0,
    "sequence_a_segment_id": 0,
    "mask_padding_with_zero": True,
}
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 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
convert_single_example Ran
ran on a synthesized input
mangushev/intent_slot/assistant/functions/intent_slot/preprocessor/preprocess.py
code served (permissive licence) · get_code("047d8e25a9cd47a0")
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 IntentSlotFeatures. 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 numpy as np

max_seq_length = 128

class PaddingInputExample(object):
    """Fake example so the num input examples is a multiple of the batch size."""
    pass

class IntentSlotFeatures(object):
    """A single set of features of data."""

    def __init__(self, input_ids, input_mask, segment_ids, intent_label_id, slot_label_ids, slot_label_mask, is_real_example=True):
        self.input_ids = input_ids
        self.input_mask = input_mask
        self.segment_ids = segment_ids
        self.intent_label_id = intent_label_id
        self.slot_label_ids = slot_label_ids
        self.slot_label_mask = slot_label_mask
        self.is_real_example = is_real_example

class InputExample(object):
    """A single training/test example for simple sequence classification."""
    def __init__(self, guid, text_a, intent_label, slot_labels):
        self.guid = guid
        self.text_a = text_a
        self.intent_label = intent_label
        self.slot_labels = slot_labels

class BasicTokenizer(object):
    def tokenize(self, text):
        return text.split()

class WordpieceTokenizer(object):
    def tokenize(self, token):
        return [token]

class SimpleTokenizer(object):
    def __init__(self):
        self.basic_tokenizer = BasicTokenizer()
        self.wordpiece_tokenizer = WordpieceTokenizer()
        self.vocab = {
            '[CLS]': 101,
            '[SEP]': 102,
            'hello': 7592,
            'world': 2088,
            'test': 3231,
            'example': 5234,
            '[PAD]': 0,
        }

    def convert_tokens_to_ids(self, tokens):
        return [self.vocab.get(token, 0) for token in tokens]

intent_label_list = ["intent_1", "intent_2", "intent_3"]
slot_label_list = ["O", "B-slot", "I-slot"]
tokenizer = SimpleTokenizer()

example = InputExample(
    guid="example_1",
    text_a="hello world test",
    intent_label="intent_1",
    slot_labels=["B-slot", "I-slot", "O"]
)

ARGS = {
    "ex_index": 0,
    "example": example,
    "intent_label_list": intent_label_list,
    "slot_label_list": slot_label_list,
    "max_seq_length": max_seq_length,
    "tokenizer": tokenizer,
}
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 IntentSlotFeatures — a IntentSlotFeatures output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a IntentSlotFeatures 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
convert_to_unicode Ran
ran; our contract draft was wrong, not the code
mangushev/intent_slot/assistant/functions/intent_slot/preprocessor/tokenization.py
pointer only (licence: MIT); the stored copy get_code serves, from huawei-noah/Pretrained-Language-Model, is pointer only (licence: NONE) · get_code("1923fc05163d207d")
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 huawei-noah/Pretrained-Language-Model, 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.

gelu Ran
ran, honoured the contract we drafted
alibaba-damo-academy/spokennlp/action-item-detection/script/modeling.py
code served (permissive licence) · get_code("d08f324f950148de")
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 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 numpy as np
import tensorflow as tf

# Create a small TensorFlow tensor for testing
x = tf.constant([[-0.5, 0.0, 0.5, 1.0], [0.2, -0.3, 0.8, -0.1]], dtype=tf.float32)

ARGS = {
    "x": x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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
get_activation Ran
ran; our contract draft was wrong, not the code
alibaba-damo-academy/spokennlp/action-item-detection/script/modeling.py
code served (permissive licence) · get_code("5842ee67106321d6")
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 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 numpy as np
import six
import tensorflow as tf

def gelu(x):
    """Gaussian Error Linear Unit.

  This is a smoother version of the RELU.
  Original paper: https://arxiv.org/abs/1606.08415
  Args:
    x: float Tensor to perform activation.

  Returns:
    `x` with the GELU activation applied.
  """
    cdf = 0.5 * (1.0 + tf.tanh(np.sqrt(2 / np.pi) * (x + 0.044715 * tf.pow(x, 3))))
    return x * cdf

def get_activation(activation_string):
    """Maps a string to a Python function, e.g., "relu" => `tf.nn.relu`.

  Args:
    activation_string: String name of the activation function.

  Returns:
    A Python function corresponding to the activation function. If
    `activation_string` is None, empty, or "linear", this will return None.
    If `activation_string` is not a string, it will return `activation_string`.

  Raises:
    ValueError: The `activation_string` does not correspond to a known
      activation.
  """
    if not isinstance(activation_string, six.string_types):
        return activation_string
    if not activation_string:
        return None
    act = activation_string.lower()
    if act == 'linear':
        return None
    elif act == 'relu':
        return tf.nn.relu
    elif act == 'gelu':
        return gelu
    elif act == 'tanh':
        return tf.tanh
    else:
        raise ValueError('Unsupported activation: %s' % act)


ARGS = {
    'activation_string': 'relu'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, six, tensorflow — 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
get_args Ran
ran; our contract draft was wrong, not the code
monologg/JointBERT/predict.py
code served (permissive licence) · get_code("107869d936646d01")
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 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 os
import tempfile
import torch

# Create a temporary directory to act as model_dir
temp_dir = tempfile.mkdtemp()

# Create a dummy training_args object to save
training_args = {
    'learning_rate': 0.001,
    'batch_size': 32,
    'epochs': 10,
    'model_name': 'test_model'
}

# Save the training_args to the temporary directory
torch.save(training_args, os.path.join(temp_dir, 'training_args.bin'))

# Create a stand-in config class with model_dir attribute
class PredConfig:
    def __init__(self, model_dir):
        self.model_dir = model_dir

pred_config = PredConfig(model_dir=temp_dir)

ARGS = {'pred_config': pred_config}
Setup block drafted by
a language model, Claude Sonnet 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 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_cn_type2video Ran
ran on a synthesized input
alibaba-damo-academy/spokennlp/mmvts/src/evaluate.py
code served (permissive licence) · get_code("c77e7567c42aa639")
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 json
import tempfile
import os

# Create a temporary JSON file with sample data
temp_dir = tempfile.mkdtemp()
cn_type2video_file = os.path.join(temp_dir, 'cn_course_0523_type2videos.json')

sample_data = {
    "type_1": ["video_1.mp4", "video_2.mp4"],
    "type_2": ["video_3.mp4", "video_4.mp4"],
    "type_3": ["video_5.mp4"]
}

with open(cn_type2video_file, 'w') as f:
    json.dump(sample_data, f)

ARGS = {
    'cn_type2video_file': cn_type2video_file
}
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 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_data_name2folder Ran
ran on a synthesized input
alibaba-damo-academy/spokennlp/emnlp2023-topic_segmentation/src/preprocess_data.py
code served (permissive licence) · get_code("28f8fe33f44d2b1c")
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 SectionProxy. 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 configparser
import tempfile
import os

# Create a temporary config file with a 'mapping' section
with tempfile.NamedTemporaryFile(mode='w', suffix='.ini', delete=False) as f:
    config_file = f.name
    f.write('[mapping]\n')
    f.write('dataset1 = /path/to/folder1\n')
    f.write('dataset2 = /path/to/folder2\n')

try:
    ARGS = {
        'config_path': config_file
    }
finally:
    pass
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 SectionProxy — a SectionProxy output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SectionProxy output the recorded ndim 1 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_device Ran
ran; our contract draft was wrong, not the code
monologg/JointBERT/predict.py
code served (permissive licence) · get_code("88395eef09dea094")
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 str. 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

class PredConfig:
    def __init__(self, no_cuda=False):
        self.no_cuda = no_cuda

pred_config = PredConfig(no_cuda=False)

ARGS = {
    'pred_config': pred_config
}
Setup block drafted by
a language model, Claude Sonnet 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 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
get_intent_labels Ran
ran on a synthesized input
monologg/JointBERT/utils.py
code served (permissive licence) · get_code("5c324b980ac8fc2f")
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 os
import tempfile

# Create a temporary directory structure with test data
temp_dir = tempfile.mkdtemp()
task_dir = os.path.join(temp_dir, "task1")
os.makedirs(task_dir, exist_ok=True)

# Create intent label file
intent_label_file = "intent_labels.txt"
intent_label_path = os.path.join(task_dir, intent_label_file)
with open(intent_label_path, 'w', encoding='utf-8') as f:
    f.write("intent_1\n")
    f.write("intent_2\n")
    f.write("intent_3\n")

class Args:
    def __init__(self, data_dir, task, intent_label_file):
        self.data_dir = data_dir
        self.task = task
        self.intent_label_file = intent_label_file

ARGS = {
    'args': Args(data_dir=temp_dir, task="task1", intent_label_file=intent_label_file)
}
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 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
get_slot_labels Ran
ran on a synthesized input
monologg/JointBERT/utils.py
code served (permissive licence) · get_code("2756ee0a98c809b9")
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 os
import tempfile

# Create a temporary directory structure with test files
temp_dir = tempfile.mkdtemp()
task_dir = os.path.join(temp_dir, "task1")
os.makedirs(task_dir, exist_ok=True)

# Create a slot label file with sample labels
slot_label_file = os.path.join(task_dir, "slot_labels.txt")
with open(slot_label_file, 'w', encoding='utf-8') as f:
    f.write("O\n")
    f.write("B-slot1\n")
    f.write("I-slot1\n")
    f.write("B-slot2\n")

# Create a simple args object
class Args:
    def __init__(self):
        self.data_dir = temp_dir
        self.task = "task1"
        self.slot_label_file = "slot_labels.txt"

ARGS = {
    "args": Args()
}
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 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
get_soft_label Ran
ran; our contract draft was wrong, not the code
dsindex/iclassifier/evaluate.py
pointer only (licence: NONE) · get_code("46fa4973164b67e8")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, dsindex/iclassifier 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.

get_topk_preds_purely Ran
ran on a synthesized input
alibaba-damo-academy/spokennlp/mmvts/src/evaluate.py
code served (permissive licence) · get_code("96d532ea3cd5bd8e")
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 int; 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

# Analyze the function:
# - labels: used with sum(labels), indexing with np.zeros_like, and list()
# - probs: used with np.argsort()
# - topk: compared with -1 and used as array index

# Create small numpy arrays for testing
labels = np.array([0, 1, 0, 1])  # 4 elements, some with sum > 0
probs = np.array([0.1, 0.9, 0.3, 0.7])  # 4 probability values
topk = 2  # positive integer topk value

ARGS = {
    "labels": labels,
    "probs": probs,
    "topk": topk
}
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 int — 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
intent_accuracy Ran
ran on a synthesized input
Polly42Rose/SiriusIntentPredictionSlotFilling/quality_metrics/quality_metrics.py
code served (permissive licence) · get_code("0933912fab50428a")
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 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

torch.manual_seed(0)

# Create small tensors for the function
# labels: shape (batch_size,) - integer class labels
# preds: shape (batch_size, num_classes) - logits/scores before argmax

batch_size = 2
num_classes = 4

labels = torch.tensor([0, 2], dtype=torch.long)
preds = torch.randn(batch_size, num_classes)

ARGS = {
    "labels": labels,
    "preds": preds,
}
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 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
load_sequences_from_file Ran
ran on a synthesized input
Polly42Rose/SiriusIntentPredictionSlotFilling/data_loading/load_data.py
code served (permissive licence) · get_code("b2e0a020c6065fe1")
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 int, shape (2, 46) 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 json
import string
import torch
import tempfile
import os

torch.manual_seed(0)

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

# Create vocabulary file
vocab = {
    "hello": 0,
    "world": 1,
    "this": 2,
    "is": 3,
    "a": 4,
    "test": 5,
    "sentence": 6,
    "with": 7,
    "words": 8
}
voc_path = os.path.join(temp_dir, "vocab.json")
with open(voc_path, 'w') as f:
    json.dump(vocab, f)

# Create data file with sample sequences
data_path = os.path.join(temp_dir, "data.txt")
with open(data_path, 'w') as f:
    f.write("hello world\n")
    f.write("this is a test\n")

ARGS = {
    "path": data_path,
    "voc_path": voc_path
}
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 int, shape (2, 46) — shape (2, 46) 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_vocab Ran
ran; our contract draft was wrong, not the code
mangushev/intent_slot/assistant/functions/intent_slot/preprocessor/tokenization.py
pointer only (licence: MIT); the stored copy get_code serves, from blcuicall/CCL2022-CLTC, is pointer only (licence: NONE) · get_code("deb4d1793cf3ba73")
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 blcuicall/CCL2022-CLTC, 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.

load_words_from_file Ran
ran on a synthesized input
Polly42Rose/SiriusIntentPredictionSlotFilling/data_loading/load_data.py
code served (permissive licence) · get_code("c807f612db31bf63")
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 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 torch
import tempfile
import os

torch.manual_seed(0)

# Create temporary files for testing
temp_dir = tempfile.mkdtemp()
path = os.path.join(temp_dir, "labels.txt")
voc_path = os.path.join(temp_dir, "vocab.json")

# Write sample labels to file
with open(path, 'w') as f:
    f.write("intent_1\n")
    f.write("intent_2\n")
    f.write("intent_1\n")
    f.write("intent_3\n")

ARGS = {
    "path": path,
    "voc_path": voc_path
}
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 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
padding_map Ran
ran on a synthesized input
Polly42Rose/SiriusIntentPredictionSlotFilling/models/dataset.py
code served (permissive licence) · get_code("680833552e4de839")
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 1, dtype family int, shape (2,) 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)

# X: tensor of shape (batch, max_seq_len) containing token indices
# The function checks X != padding_value, so we need a tensor with some padding tokens
X = torch.tensor([
    [1, 2, 3, 0],      # batch 0: 3 non-padding tokens
    [5, 6, 0, 0]       # batch 1: 2 non-padding tokens
])

# padding_value: the token ID used for padding
padding_value = 0

ARGS = {
    "X": X,
    "padding_value": padding_value
}
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 1, dtype family int, shape (2,) — shape (2,) 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
printable_text Ran
ran; our contract draft was wrong, not the code
mangushev/intent_slot/assistant/functions/intent_slot/preprocessor/tokenization.py
code served (permissive licence) · get_code("0e5615f8994003cf")
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 six

# The function takes a single parameter 'text' which can be:
# - str (works in both Python 2 and 3)
# - bytes (in Python 3, decoded to utf-8)
# - unicode (in Python 2, encoded to utf-8)

# Since we're running on Python 3, we'll test with a string
text_input = "Hello, World!"

ARGS = {
    "text": text_input
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
six — 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
read_file Ran
ran; our contract draft was wrong, not the code
yinghao1019/Joint_learn/predict.py
pointer only (licence: NONE) · get_code("7b9905a68c0add30")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yinghao1019/Joint_learn 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.

read_jsonl_file Ran
ran on a synthesized input
alibaba-damo-academy/spokennlp/mmvts/src/evaluate.py
code served (permissive licence) · get_code("90bf04bed34edf57")
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 json
import tempfile
import os

# Create a temporary JSONL file with sample data
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test_data.jsonl")

# Write sample JSONL data
sample_data = [
    {"id": 1, "text": "hello world", "value": 42},
    {"id": 2, "text": "foo bar", "value": 100},
    {"id": 3, "text": "test data", "value": 7}
]

with open(temp_file, 'w') as f:
    for record in sample_data:
        f.write(json.dumps(record) + "\n")

ARGS = {
    "in_file": temp_file
}
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 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
read_total_pred_and_labels Ran
ran on a synthesized input
alibaba-damo-academy/spokennlp/emnlp2023-topic_segmentation/src/postprocess_predictions.py
code served (permissive licence) · get_code("e9b0cd55fca10310")
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. 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 json
import tempfile
import os

# Create temporary files with the required JSON format
pred_file_content = [
    json.dumps({'labels': ['O', 'O', 'B-PER', 'O'], 'predictions': ['O', 'B-PER', 'B-PER', 'O']}),
    json.dumps({'labels': ['B-LOC', 'O', 'O'], 'predictions': ['B-LOC', 'O', 'B-LOC']})
]

data_file_content = [
    json.dumps({'labels': [0, 1, 0, 1, 0]}),
    json.dumps({'labels': [1, 1, 0, 0]})
]

# Create temporary files
with tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.jsonl') as pred_f:
    pred_file_path = pred_f.name
    for line in pred_file_content:
        pred_f.write(line + '\n')

with tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.jsonl') as data_f:
    data_file_path = data_f.name
    for line in data_file_content:
        data_f.write(line + '\n')

ARGS = {
    'data_file': data_file_path,
    'pred_file': pred_file_path
}
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 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
sentence_accuracy Ran
ran on a synthesized input
Polly42Rose/SiriusIntentPredictionSlotFilling/quality_metrics/quality_metrics.py
code served (permissive licence) · get_code("0a3ade9a74c9987d")
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 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

torch.manual_seed(0)

# Analyze the function:
# - i_labels: indexed and compared with i_preds[i] (should be 1D tensor of ints)
# - i_preds: argmax applied on dim=1, then indexed with i (should be 2D tensor)
# - s_labels: indexed with i, then indexed with j, compared with s_preds[i][j] (should be 2D tensor)
# - s_preds: argmax applied on dim=2, then indexed with i and j (should be 3D tensor)

batch_size = 2
num_classes = 4
seq_len = 3
num_seq_classes = 5

# i_labels: shape (batch_size,) - 1D tensor of class labels
i_labels = torch.tensor([0, 1])

# i_preds: shape (batch_size, num_classes) - 2D tensor of logits for argmax on dim=1
i_preds = torch.randn(batch_size, num_classes)

# s_labels: shape (batch_size, seq_len) - 2D tensor of sequence labels
s_labels = torch.tensor([[0, 1, 2], [1, 2, 0]])

# s_preds: shape (batch_size, seq_len, num_seq_classes) - 3D tensor of logits for argmax on dim=2
s_preds = torch.randn(batch_size, seq_len, num_seq_classes)

ARGS = {
    "i_labels": i_labels,
    "i_preds": i_preds,
    "s_labels": s_labels,
    "s_preds": s_preds,
}
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 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
set_device Ran
ran; our contract draft was wrong, not the code
yinghao1019/Joint_learn/predict.py
pointer only (licence: NONE) · get_code("cfd666d38b923d90")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yinghao1019/Joint_learn 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.

slot_f1 Ran
ran on a synthesized input
Polly42Rose/SiriusIntentPredictionSlotFilling/quality_metrics/quality_metrics.py
code served (permissive licence) · get_code("b62b0764cd5a1413")
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, 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 torch
import numpy as np

torch.manual_seed(0)

# The function expects:
# - labels: a 2D tensor of shape (batch_size, sequence_length) with integer slot indices
# - preds: a 3D tensor of shape (batch_size, sequence_length, num_slot) with logits/scores
# - num_slot: integer number of slots

batch_size = 2
sequence_length = 4
num_slot = 3

# Create labels: shape (batch_size, sequence_length) with values in [0, num_slot)
labels = torch.randint(0, num_slot, (batch_size, sequence_length))

# Create preds: shape (batch_size, sequence_length, num_slot) with random logits
# argmax will be applied to dim=2, so we need proper shape
preds = torch.randn(batch_size, sequence_length, num_slot)

ARGS = {
    "labels": labels,
    "preds": preds,
    "num_slot": num_slot
}
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, ndim 1, dtype family float — a tuple 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
file_based_input_fn_builder Not yet run
unverified (harvested, no recorded run)
mangushev/intent_slot/bert/run_intent_slot.py
code served (permissive licence) · get_code("6c079a61fb350572")
How it ran

no invocation recorded.

get_assignment_map_from_checkpoint Not yet run
unverified (harvested, no recorded run)
alibaba-damo-academy/spokennlp/action-item-detection/script/modeling.py
code served (permissive licence) · get_code("a8a77945aa9b653f")
How it ran

no invocation recorded.

load_data Not yet run
unverified (harvested, no recorded run)
Polly42Rose/SiriusIntentPredictionSlotFilling/data_loading/load_data.py
code served (permissive licence) · get_code("96f7ae2ec61191ba")
How it ran

no invocation recorded.

preprocess_atis Not yet run
unverified (harvested, no recorded run)
Polly42Rose/SiriusIntentPredictionSlotFilling/models/dataset.py
code served (permissive licence) · get_code("89bccc2b443e9266")
How it ran

no invocation recorded.

process_wiki_section_subset Not yet run
unverified (harvested, no recorded run)
alibaba-damo-academy/spokennlp/emnlp2023-topic_segmentation/src/preprocess_data.py
code served (permissive licence) · get_code("40b50b726e7a6434")
How it ran

no invocation recorded.

tokenize_method Not yet run
unverified (harvested, no recorded run)
alibaba-damo-academy/spokennlp/emnlp2023-topic_segmentation/src/preprocess_data.py
code served (permissive licence) · get_code("e83d18c25c1278b9")
How it ran

no invocation recorded.

Also present in other papers' code

Of 32 samples lifted from this paper's linked repositories, 32 carry a function identity and 0 do not. Of the 32 distinct functions with an identity, 9 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 4 the same month, 7 later) and 23 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
convert_to_unicode 36 other papers
4 dated earlier, 1 the same month, 31 later
1609.08144 (2016-09)
Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1901.04085 (2019-01)
Passage Re-ranking with BERT
and 33 more not named here
printable_text 33 other papers
4 dated earlier, 1 the same month, 28 later
1609.08144 (2016-09)
Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1901.04085 (2019-01)
Passage Re-ranking with BERT
and 30 more not named here
get_activation 32 other papers
4 dated earlier, 1 the same month, 27 later · 1 only as an assembly brick
1804.05392 (2018-04)
Higher-order Coreference Resolution with Coarse-to-fine Inference
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1901.04085 (2019-01)
Passage Re-ranking with BERT
and 29 more not named here
gelu 26 other papers
3 dated earlier, 2 the same month, 21 later · 4 only as an assembly brick
1706.03762 (2017-06)
Attention Is All You Need
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1901.02860 (2019-01)
Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
and 23 more not named here
load_vocab 4 other papers
4 later
1903.09588 (2019-03)
Utilizing BERT for Aspect-Based Sentiment Analysis via Constructing Auxiliary Sentence
2203.09313 (2022-03)
EVA2.0: Investigating Open-Domain Chinese Dialogue Systems with Large-Scale Pre-Training
2310.18738 (2023-10)
TLM: Token-Level Masking for Transformers
and 1 more not named here
get_args 1 other paper
1 later
1905.08284 (2019-05)
Enriching Pre-trained Language Model with Entity Information for Relation Classification
get_device 1 other paper
1 later
1905.08284 (2019-05)
Enriching Pre-trained Language Model with Entity Information for Relation Classification
read_file 1 other paper
1 dated earlier
1609.01454 (2016-09)
Attention-Based Recurrent Neural Network Models for Joint Intent Detection and Slot Filling
set_device 1 other paper
1 dated earlier
1609.01454 (2016-09)
Attention-Based Recurrent Neural Network Models for Joint Intent Detection and Slot Filling

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

Intent classification and slot filling are two essential tasks for natural language understanding. They often suffer from small-scale human-labeled training data, resulting in poor generalization capability, especially for rare words. Recently a new language representation model, BERT (Bidirectional Encoder Representations from Transformers), facilitates pre-training deep bidirectional representations on large-scale unlabeled corpora, and has created state-of-the-art models for a wide variety of natural language processing tasks after simple fine-tuning. However, there has not been much effort on exploring BERT for natural language understanding. In this work, we propose a joint intent classification and slot filling model based on BERT. Experimental results demonstrate that our proposed model achieves significant improvement on intent classification accuracy, slot filling F1, and sentence-level semantic frame accuracy on several public benchmark datasets, compared to the attention-based recurrent neural network models and slot-gated models.

For agents

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

get_harvested_code_for_paper("1902.10909")
get_code_for_paper("1902.10909")
have("1902.10909")

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

curl https://syntology.ai/api/ran/1902.10909.json

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

[![Syntology run record](https://syntology.ai/api/ran/1902.10909.svg)](https://syntology.ai/paper/1902.10909)

Connect an agent — have() is free.