SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2112.01640 · 2021

MultiVerS: Improving scientific claim verification with weak supervision and full-document context

arXiv · PDF · Open in the Atlas

Code that ran

13 of 13 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, 1 violated it and 12 have no contract verdict; 0 failures of our instrument; 0 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/2112.01640.json · badge: /api/ran/2112.01640.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
dwadden/longchecker canonical (classifier label)
classifier basis not recorded on the link
13 ran · 0 no recorded run
FunctionStatusWhere it lives
batch_tensor_dicts Ran
ran on a synthesized input
dwadden/longchecker/multivers/allennlp_nn_util.py
code served (permissive licence) · get_code("d0f21f4a3c8edaca")
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
from collections import defaultdict

torch.manual_seed(0)

# Create a list of tensor dictionaries to batch
# Each dictionary has the same keys but different tensor values
tensor_dicts = [
    {
        "embeddings": torch.randn(1, 4),
        "logits": torch.randn(1, 3),
        "attention": torch.randn(1, 2)
    },
    {
        "embeddings": torch.randn(1, 4),
        "logits": torch.randn(1, 3),
        "attention": torch.randn(1, 2)
    }
]

ARGS = {
    "tensor_dicts": tensor_dicts,
    "remove_trailing_dimension": True
}
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
clamp_tensor Ran
ran on a synthesized input
dwadden/longchecker/multivers/allennlp_nn_util.py
code served (permissive licence) · get_code("3d52bb92a6570116")
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. 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)

# Create a sparse tensor
indices = torch.tensor([[0, 1, 2], [0, 1, 2]])
values = torch.tensor([1.5, 2.5, 3.5])
sparse_tensor = torch.sparse_coo_tensor(indices, values, (3, 3))

# Create a dense tensor
dense_tensor = torch.tensor([[1.5, 2.5], [3.5, 4.5]])

ARGS = {
    "tensor": sparse_tensor,
    "minimum": 2.0,
    "maximum": 3.0
}
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, shape (3, 3) — shape (3, 3) as a later run recorded it
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
compute_f1 Ran
ran on a synthesized input
dwadden/longchecker/multivers/metrics.py
code served (permissive licence) · get_code("7e52643dedb5d9d4")
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

torch.manual_seed(0)

# The function compute_f1 takes 4 parameters: relevant, retrieved, correct, prefix
# All are used in arithmetic operations (safe_divide) and string formatting
# They should be numeric values and a string

ARGS = {
    'relevant': 10,
    'retrieved': 8,
    'correct': 6,
    'prefix': 'test'
}
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
flatten Ran
ran on a synthesized input
dwadden/longchecker/multivers/util.py
code served (permissive licence) · get_code("5737801552204b46")
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
import numpy as np

# The flatten function takes a nested list (z) and flattens it
# It iterates through each element y in z, then through each x in y
# So z should be a list of lists

ARGS = {
    "z": [[1, 2, 3], [4, 5], [6, 7, 8, 9]]
}
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 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_tokenizer Ran
ran on a synthesized input
dwadden/longchecker/multivers/data_train.py
code served (permissive licence) · get_code("b49b5d0c59c33db8")
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 MockTokenizer. 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
from unittest.mock import MagicMock

class MockHParams:
    def __init__(self, encoder_name):
        self.encoder_name = encoder_name

class MockTokenizer:
    def __init__(self):
        self.vocab_size = 50000
        self.added_tokens = []
    
    def add_tokens(self, tokens):
        """Mock method to add tokens to tokenizer."""
        self.added_tokens.extend(tokens)
    
    def __call__(self, text, **kwargs):
        """Mock tokenization."""
        return {"input_ids": [1, 2, 3], "attention_mask": [1, 1, 1]}

# Create a mock for AutoTokenizer.from_pretrained
def mock_from_pretrained(model_name):
    return MockTokenizer()

# Monkey-patch AutoTokenizer for testing
import transformers
transformers.AutoTokenizer.from_pretrained = staticmethod(mock_from_pretrained)

# Test case 1: Non-special encoder name
hparams_standard = MockHParams(encoder_name='bert-base-uncased')

# Test case 2: Special longformer encoder name (triggers additional tokens logic)
hparams_longformer = MockHParams(encoder_name='longformer-large-science')

ARGS = {
    'hparams': hparams_standard
}
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 MockTokenizer — a MockTokenizer output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a MockTokenizer 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
has_correct_rationale_abstract Ran
ran on a synthesized input
dwadden/longchecker/multivers/metrics.py
code served (permissive licence) · get_code("eda2fde7bfd6ee5f")
How it ran

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

import torch

torch.manual_seed(0)

# Construct valid arguments for has_correct_rationale_abstract
# The function takes two parameters: pred and gold, both tensors

# pred: a tensor of predictions (binary indicators of predicted rationales)
# gold: a tensor of gold labels (sentence IDs for rationales, with 0 meaning non-rationale)

# Create a small pred tensor (binary predictions)
pred = torch.tensor([1, 0, 1, 0], dtype=torch.float32)

# Create a small gold tensor with gold rationale IDs
# Values > 0 indicate the sentence ID of a gold rationale
# Value 0 means not a rationale
gold = torch.tensor([1, 0, 1, 2], dtype=torch.long)

ARGS = {
    "pred": pred,
    "gold": gold,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, ndim 0, dtype family int, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
has_tensor Ran
ran, violated the contract we drafted
dwadden/longchecker/multivers/allennlp_nn_util.py
code served (permissive licence) · get_code("c71aafcb215c5122")
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 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.

import torch

torch.manual_seed(0)

# Test case 1: A simple tensor
obj1 = torch.randn(2, 4)

# Test case 2: A dict containing a tensor
obj2 = {
    "tensor": torch.randn(2, 4),
    "value": 42
}

# Test case 3: A list containing tensors
obj3 = [torch.randn(2, 4), "string", 123]

# Test case 4: A tuple containing tensors
obj4 = (torch.randn(2, 4), "data")

# Test case 5: Nested structure with tensors deep inside
obj5 = {
    "nested": {
        "list": [torch.randn(2, 4), 100],
        "value": "test"
    },
    "other": 999
}

# Test case 6: No tensors at all
obj6 = {
    "a": 1,
    "b": [2, 3, "four"],
    "c": (5, 6)
}

# Test case 7: Empty structures
obj7 = []

# Test case 8: Complex nested without tensors
obj8 = {"x": [1, 2], "y": (3, 4), "z": {"nested": "value"}}

ARGS = {
    "obj": obj1
}
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 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_jsonl Ran
ran on a synthesized input
dwadden/longchecker/multivers/data_verisci.py
code served (permissive licence) · get_code("03dab1e2aeebd1fc")
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()
fname = os.path.join(temp_dir, "test.jsonl")

# Write some valid JSON lines to the file
with open(fname, 'w') as f:
    f.write(json.dumps({"key1": "value1", "key2": 42}) + '\n')
    f.write(json.dumps({"key1": "value2", "key2": 43}) + '\n')
    f.write(json.dumps({"key1": "value3", "key2": 44}) + '\n')

ARGS = {
    "fname": fname
}
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
load_jsonl Ran
ran on a synthesized input
dwadden/longchecker/multivers/util.py
code served (permissive licence) · get_code("e6ce4441929bac8d")
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_fd, temp_path = tempfile.mkstemp(suffix='.jsonl')
try:
    with os.fdopen(temp_fd, 'w') as f:
        f.write(json.dumps({"key": "value1"}) + "\n")
        f.write(json.dumps({"key": "value2"}) + "\n")
        f.write(json.dumps({"key": "value3"}) + "\n")
        f.write(json.dumps({"key": "value4"}) + "\n")
    
    ARGS = {
        "fname": temp_path,
        "max_lines": 2
    }
except:
    os.unlink(temp_path)
    raise
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
make_label Ran
ran on a synthesized input
dwadden/longchecker/multivers/data_verisci.py
code served (permissive licence) · get_code("f5065669b044e99f")
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 Label. 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
from enum import Enum

class Label(Enum):
    SUPPORTS = 2
    NEI = 1
    REFUTES = 0

def make_label(label_str, allow_NEI=True):
    lookup = {'SUPPORT': Label.SUPPORTS, 'NOT_ENOUGH_INFO': Label.NEI, 'CONTRADICT': Label.REFUTES}
    res = lookup[label_str]
    if not allow_NEI and res is Label.NEI:
        raise ValueError('An NEI was given.')
    return res

ARGS = {
    'label_str': 'SUPPORT',
    'allow_NEI': True
}
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 Label — a Label output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Label 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
masked_binary_cross_entropy_with_logits Ran
ran on a synthesized input
dwadden/longchecker/multivers/model.py
code served (permissive licence) · get_code("7810f9d40d608c62")
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 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 torch.nn.functional as F

torch.manual_seed(0)

# Create small tensors for the function arguments
batch_size = 2
seq_length = 4

# input: logits tensor, shape (batch_size, seq_length)
input_tensor = torch.randn(batch_size, seq_length)

# target: binary targets with -1 for ignore, shape (batch_size, seq_length)
target_tensor = torch.tensor([
    [1, 0, 1, -1],
    [0, 1, -1, 1]
], dtype=torch.long)

# weight: scalar weight for each sample, shape (batch_size,)
weight_tensor = torch.tensor([1.0, 0.5])

# rationale_mask: binary mask, shape (batch_size,)
rationale_mask_tensor = torch.tensor([1.0, 1.0])

ARGS = {
    "input": input_tensor,
    "target": target_tensor,
    "weight": weight_tensor,
    "rationale_mask": rationale_mask_tensor
}
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 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
safe_divide Ran
ran on a synthesized input
dwadden/longchecker/multivers/metrics.py
code served (permissive licence) · get_code("b999fdcde5c96886")
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)

ARGS = {
    "num": 10,
    "denom": 2,
}
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
unbatch Ran
ran on a synthesized input
dwadden/longchecker/multivers/util.py
code served (permissive licence) · get_code("89a74da47ab6bbe4")
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 torch

torch.manual_seed(0)

# Create a batched dictionary with tensors of batch size 2
d = {
    "tensor_1": torch.randn(2, 4),
    "tensor_2": torch.randn(2, 3),
    "non_tensor": [1, 2],  # non-tensor value with length 2
}

ignore = ["non_tensor"]

ARGS = {
    "d": d,
    "ignore": ignore,
}
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 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

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 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 10 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
has_tensor 1 other paper
1 later
2310.18685 (2023-10)
When Reviewers Lock Horn: Finding Disagreement in Scientific Peer Reviews
load_jsonl 1 other paper
1 later
2023.findings-acl.888 (2023 (year only))
Check-COVID: Fact-Checking COVID-19 News Claims with Scientific Evidence
masked_binary_cross_entropy_with_logits 1 other paper
1 later
2301.13298 (2023-01)
LongEval: Guidelines for Human Evaluation of Faithfulness in Long-form Summarization

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

The scientific claim verification task requires an NLP system to label scientific documents which Support or Refute an input claim, and to select evidentiary sentences (or rationales) justifying each predicted label. In this work, we present MultiVerS, which predicts a fact-checking label and identifies rationales in a multitask fashion based on a shared encoding of the claim and full document context. This approach accomplishes two key modeling goals. First, it ensures that all relevant contextual information is incorporated into each labeling decision. Second, it enables the model to learn from instances annotated with a document-level fact-checking label, but lacking sentence-level rationales. This allows MultiVerS to perform weakly-supervised domain adaptation by training on scientific documents labeled using high-precision heuristics. Our approach outperforms two competitive baselines on three scientific claim verification datasets, with particularly strong performance in zero / few-shot domain adaptation experiments. Our code and data are available at https://github.com/dwadden/multivers.

For agents

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

get_harvested_code_for_paper("2112.01640")
get_code_for_paper("2112.01640")
have("2112.01640")

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

curl https://syntology.ai/api/ran/2112.01640.json

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

[![Syntology run record](https://syntology.ai/api/ran/2112.01640.svg)](https://syntology.ai/paper/2112.01640)

Connect an agent — have() is free.