SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1804.07754 · 2018

Learning Semantic Textual Similarity from Conversations

arXiv · PDF · Open in the Atlas

Code that ran

15 of 17 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; 0 honoured the contract we drafted, 0 violated it and 15 have no contract verdict; 0 failures of our instrument; 2 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/1804.07754.json · badge: /api/ran/1804.07754.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
nickyeolk/info_retrieve community (classifier label)
classifier basis not recorded on the link
15 ran · 2 no recorded run
FunctionStatusWhere it lives
add_period Ran
ran on a synthesized input
nickyeolk/info_retrieve/src/importance.py
code served (permissive licence) · get_code("fc35b5b7943a910b")
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 string

def add_period(word):
    if word in string.punctuation:
        return word
    else:
        return ' ' + word

# The function takes a single parameter 'word' which is a string
# It checks if the word is in string.punctuation
# If not, it prepends a space to the word

ARGS = {
    'word': 'hello'
}
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
add_period_html Ran
ran on a synthesized input
nickyeolk/info_retrieve/src/importance.py
code served (permissive licence) · get_code("6fc5d2d5b1f9523e")
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 re
import string

def add_period(word):
    if word in string.punctuation:
        return word
    else:
        return ' ' + word

def add_period_html(word):
    if ('>' in word) & ('<' in word):
        if re.findall('>(.*?)<', word)[0] in string.punctuation:
            return word
        else:
            return ' ' + word
    else:
        return add_period(word)

# Test case 1: HTML with punctuation inside tags
test_word_html_punct = ">.<"

ARGS = {
    "word": test_word_html_punct
}
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
clean_txt Ran
ran on a synthesized input
nickyeolk/info_retrieve/src/utils.py
code served (permissive licence) · get_code("6d48135dd9946bb4")
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

def clean_txt(text):
    """
    Strips formatting
    """
    if isinstance(text, str):
        return text.replace('\n', '. ').replace('.. ', '. ').rstrip()
    elif hasattr(text, '__iter__'):
        return [clean_txt(one_text) for one_text in text]

torch.manual_seed(0)

# Test case 1: simple string
test_str = "Hello\nWorld\n.. test"

# Test case 2: list of strings
test_list = ["First\nline", "Second.. line", "Third\n\nline"]

# The function accepts either a string or an iterable of strings
ARGS = {
    "text": test_str
}
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
extract Ran
ran on a synthesized input
nickyeolk/info_retrieve/app_flask.py
code served (permissive licence) · get_code("a81e3c623f9e5618")
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 2, 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

# The function extract(lst, idx=0) takes a list and an index,
# and returns a list where each element is item[idx] from the original list.
# So lst should be a list of indexable items (lists, tuples, tensors, etc.)

torch.manual_seed(0)

# Create a list of tuples/lists where each item can be indexed
lst = [
    (torch.tensor([1.0, 2.0, 3.0, 4.0]), torch.tensor([5.0, 6.0])),
    (torch.tensor([7.0, 8.0, 9.0, 10.0]), torch.tensor([11.0, 12.0])),
]

idx = 0

ARGS = {
    "lst": lst,
    "idx": idx,
}
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 list, ndim 2, 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
format_func Ran
ran on a synthesized input
nickyeolk/info_retrieve/st_demo.py
code served (permissive licence) · get_code("d4c5fb19a0477c97")
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 = {
    'kb_name': 'covid19'
}
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
generate_mappings Ran
ran on a synthesized input
nickyeolk/info_retrieve/src/kb_handler.py
code served (permissive licence) · get_code("cd246cb6c6617d87")
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 2, 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 pandas as pd

# Create sample pd.Series for responses and queries with some duplicates
responses = pd.Series(['response_a', 'response_b', 'response_a', 'response_c'])
queries = pd.Series(['query_x', 'query_y', 'query_x', 'query_z'])

ARGS = {
    'responses': responses,
    'queries': queries
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 2, 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
get_ids Ran
ran on a synthesized input
nickyeolk/info_retrieve/src/bert_tokenization.py
code served (permissive licence) · get_code("6b4945d2e09fe154")
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 torch
from transformers import AutoTokenizer

# Create a minimal tokenizer stand-in that implements convert_tokens_to_ids
class MockTokenizer:
    def __init__(self):
        # Simple vocab mapping
        self.vocab = {
            "[PAD]": 0,
            "hello": 1,
            "world": 2,
            "test": 3,
            "token": 4,
            "[UNK]": 5,
        }
    
    def convert_tokens_to_ids(self, tokens):
        """Convert tokens to their corresponding ids"""
        return [self.vocab.get(token, self.vocab["[UNK]"]) for token in tokens]

# Create test data
tokens = ["hello", "world", "test"]
tokenizer = MockTokenizer()
max_seq_length = 8

ARGS = {
    "tokens": tokens,
    "tokenizer": tokenizer,
    "max_seq_length": max_seq_length,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, transformers — 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
get_masks Ran
ran on a synthesized input
nickyeolk/info_retrieve/src/bert_tokenization.py
code served (permissive licence) · get_code("fe26978879cca681")
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 torch

torch.manual_seed(0)

# tokens parameter: a list of token IDs
tokens = [101, 2054, 2003, 1045]

# max_seq_length parameter: an integer specifying maximum sequence length
max_seq_length = 8

ARGS = {
    "tokens": tokens,
    "max_seq_length": max_seq_length
}
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, 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
get_segments Ran
ran on a synthesized input
nickyeolk/info_retrieve/src/bert_tokenization.py
code served (permissive licence) · get_code("922a8461783d514b")
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 torch

torch.manual_seed(0)

# The function takes a list of tokens and a max_seq_length
# It returns a list of segment IDs (0 or 1)
# Segments are 0 for first sequence, 1 for second sequence after [SEP]

tokens = ['[CLS]', 'hello', 'world', '[SEP]', 'foo', 'bar']
max_seq_length = 10

ARGS = {
    'tokens': tokens,
    'max_seq_length': max_seq_length
}
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, 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
mrr Ran
ran on a synthesized input
nickyeolk/info_retrieve/src/finetune_eval.py
code served (permissive licence) · get_code("7c606cc27a65fdbb")
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
import numpy as np

# The mrr function takes a list of ranks (predicted positions of correct responses)
# and calculates the mean reciprocal rank by summing 1/rank for each rank and dividing by count

# Create a small list of ranks
# Ranks should be positive integers representing positions (1st, 2nd, 3rd, etc.)
ranks = [1, 2, 5, 3]

ARGS = {
    "ranks": ranks
}
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 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
parse_string Ran
ran on a synthesized input
nickyeolk/info_retrieve/src/importance.py
code served (permissive licence) · get_code("7041cf85b5729cb5")
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 re
import string

def parse_string(string):
    """
    Parse sentence string into:
        1. its words and punctuations
        2. List of idx
    """
    list_of_words_and_puncs = re.findall('[\\w\']+|[.,!?;-~-{}`´_<=>:/@*()&\'$%#"]', string)
    list_of_word_idx = [i for i, word in enumerate(list_of_words_and_puncs)]
    return list_of_words_and_puncs


ARGS = {
    "string": "Hello, world! This is a test."
}
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
question_cleaner Ran
ran on a synthesized input
nickyeolk/info_retrieve/src/utils.py
code served (permissive licence) · get_code("c8a48beb0be3672c")
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 DataFrame. 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 pandas as pd
import numpy as np

# Create a sample DataFrame that mimics the structure expected by question_cleaner
# The function expects columns at indices 2 and 3
# Column 2 (gt): ground truth answers as space-separated integers
# Column 3 (kb): knowledge base answers as space-separated integers

data = {
    0: ['question1', 'question2'],
    1: ['context1', 'context2'],
    2: ['1 2 3', '4 5'],  # ground truth answers (space-separated integers)
    3: ['2 3 4', '6 7']   # knowledge base answers (space-separated integers)
}

df_query = pd.DataFrame(data)

ARGS = {
    'df_query': df_query
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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_txt Ran
ran on a synthesized input
nickyeolk/info_retrieve/src/utils.py
code served (permissive licence) · get_code("9aff0e1769d7891e")
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 tempfile
import os

# Create a temporary file with some test content
temp_dir = tempfile.mkdtemp()
test_file_path = os.path.join(temp_dir, "test.txt")

with open(test_file_path, 'w', encoding='utf-8') as f:
    f.write("Line 1: First clause\n")
    f.write("Line 2: Second clause\n")
    f.write("Line 3: Third clause\n")

ARGS = {
    "path": test_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 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
triplet_loss Ran
ran on a synthesized input
nickyeolk/info_retrieve/src/metric_learning.py
code served (permissive licence) · get_code("370a6e91bad59927")
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 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 tensorflow as tf
import numpy as np

# Set seed for reproducibility
tf.random.set_seed(0)
np.random.seed(0)

# Create small embedding vectors (batch_size=2, embedding_dim=4)
batch_size = 2
embedding_dim = 4

# Anchor vectors: shape (2, 4)
anchor_vector = tf.constant([
    [0.1, 0.2, 0.3, 0.4],
    [0.5, 0.6, 0.7, 0.8]
], dtype=tf.float32)

# Positive vectors: shape (2, 4)
positive_vector = tf.constant([
    [0.11, 0.21, 0.31, 0.41],
    [0.51, 0.61, 0.71, 0.81]
], dtype=tf.float32)

# Negative vectors: shape (2, 4)
negative_vector = tf.constant([
    [0.9, 0.8, 0.7, 0.6],
    [0.5, 0.4, 0.3, 0.2]
], dtype=tf.float32)

ARGS = {
    'anchor_vector': anchor_vector,
    'positive_vector': positive_vector,
    'negative_vector': negative_vector,
    'metric': 'cosine_dist',
    'margin': 0.009
}
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 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
unique_indexing Ran
ran on a synthesized input
nickyeolk/info_retrieve/src/kb_handler.py
code served (permissive licence) · get_code("68b251e42b81afcd")
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 pandas as pd

ARGS = {
    "non_unique": pd.Series(["apple", "banana", "apple", "cherry", "banana", "apple"])
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — 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
hard_triplet_generator Not yet run
unverified (harvested, no recorded run)
nickyeolk/info_retrieve/src/finetune_eval.py
code served (permissive licence) · get_code("143be1a6554ab483")
How it ran

no invocation recorded.

random_triplet_generator Not yet run
unverified (harvested, no recorded run)
nickyeolk/info_retrieve/src/finetune_eval.py
code served (permissive licence) · get_code("d1ccb61839e8ab51")
How it ran

no invocation recorded.

Also present in other papers' code

Of 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 17 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 17 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.

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

We present a novel approach to learn representations for sentence-level semantic similarity using conversational data. Our method trains an unsupervised model to predict conversational input-response pairs. The resulting sentence embeddings perform well on the semantic textual similarity (STS) benchmark and SemEval 2017's Community Question Answering (CQA) question similarity subtask. Performance is further improved by introducing multitask training combining the conversational input-response prediction task and a natural language inference task. Extensive experiments show the proposed model achieves the best performance among all neural models on the STS benchmark and is competitive with the state-of-the-art feature engineered and mixed systems in both tasks.

For agents

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

get_harvested_code_for_paper("1804.07754")
get_code_for_paper("1804.07754")
have("1804.07754")

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

curl https://syntology.ai/api/ran/1804.07754.json

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

[![Syntology run record](https://syntology.ai/api/ran/1804.07754.svg)](https://syntology.ai/paper/1804.07754)

Connect an agent — have() is free.