SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.10426 · 2024

DELL: Generating Reactions and Explanations for LLM-Based Misinformation Detection

arXiv · PDF · Open in the Atlas

Code that ran

6 of 7 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 0 have no contract verdict; 6 failures of our instrument; 1 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/2402.10426.json · badge: /api/ran/2402.10426.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
whr000001/dell canonical (classifier label)
classifier basis: census
6 ran · 1 no recorded run
FunctionStatusWhere it lives
find_bracketed_substrings Ran
ran; our contract draft was wrong, not the code
whr000001/dell/ensemble/confidence.py
code served (permissive licence) · get_code("d8bb61a0d7699784")
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

def find_bracketed_substrings(input_string):
    pattern = '\\[(.*?)\\]'
    res = re.findall(pattern, input_string)
    return res[0]

# The function takes a string and finds all content within square brackets
# It returns the first match found by the regex pattern
# We need to provide a string that contains at least one bracketed substring

ARGS = {
    "input_string": "Hello [world] this is a test [example]"
}
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
get_metric Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
whr000001/dell/ensemble/confidence.py
code served (permissive licence) · get_code("55aa60d550a7a519")
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
from sklearn.metrics import f1_score

torch.manual_seed(0)

# Create simple ground truth and predictions for multi-class classification
# Using 3 classes, 5 samples
y_true = np.array([0, 1, 2, 1, 0])
y_pred = np.array([0, 1, 1, 1, 0])

ARGS = {
    "y_true": y_true,
    "y_pred": y_pred
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn, 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
get_metric Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
whr000001/dell/ensemble/ensemble_utils.py
code served (permissive licence) · get_code("69b987d6d189b093")
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 numpy as np
from sklearn.metrics import f1_score

# The function takes y_true and y_pred arguments
# y_true: ground truth labels
# y_pred: predicted labels
# Both should be array-like with class labels for multi-class classification

# Create small example data for multi-class classification
# Using 3 classes with 6 samples
y_true_data = np.array([0, 1, 2, 0, 1, 2])
y_pred_data = np.array([0, 1, 1, 0, 1, 2])

ARGS = {
    'y_true': y_true_data,
    'y_pred': y_pred_data,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn — 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
multilabel_categorical_cross_entropy Ran
ran; our contract draft was wrong, not the code
whr000001/dell/expert/model.py
code served (permissive licence) · get_code("7748d4b6529cccf3")
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

torch.manual_seed(0)

# y_pred: predictions tensor, shape (batch_size, num_labels)
# y_true: binary labels tensor, shape (batch_size, num_labels)
# Both should have same shape and contain float values

batch_size = 2
num_labels = 4

y_pred = torch.randn(batch_size, num_labels, dtype=torch.float32)
y_true = torch.randint(0, 2, (batch_size, num_labels), dtype=torch.float32)

ARGS = {
    "y_pred": y_pred,
    "y_true": y_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 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
my_collate_fn Ran
ran; our contract draft was wrong, not the code
whr000001/dell/expert/dataset.py
code served (permissive licence) · get_code("4a389827402efacf")
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)

# Construct a batch of items with all required keys
batch = [
    {
        'text': 'This is a sample text.',
        'sentiment': 0,
        'framing': 1,
        'propaganda': 0,
        'retrieval': [0.1, 0.2, 0.3, 0.4],
        'label': 1,
        'graph_info': {'nodes': 5, 'edges': 3}
    },
    {
        'text': 'Another sample text here.',
        'sentiment': 1,
        'framing': 0,
        'propaganda': 1,
        'retrieval': [0.5, 0.6, 0.7, 0.8],
        'label': 0,
        'graph_info': {'nodes': 7, 'edges': 4}
    }
]

ARGS = {
    'batch': batch
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
validation Ran
ran; our contract draft was wrong, not the code
whr000001/dell/expert/inference.py
code served (permissive licence) · get_code("4d861dde8c1140a8")
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 torch
import torch.nn as nn
from torch.utils.data import DataLoader

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, batch):
        # batch is a dict with 'label' key
        # Return (out, _) where out is logits
        out = self.linear(batch['input'])
        return out, None

class SimpleDataset:
    def __init__(self, num_samples=2, task='TASK1'):
        self.num_samples = num_samples
        self.task = task
    
    def __iter__(self):
        for i in range(self.num_samples):
            if self.task == 'TASK1':
                # Multi-class: labels in range [0, num_classes)
                label = torch.tensor([i % 3], dtype=torch.long)
            else:
                # Multi-label: binary labels
                label = torch.tensor([i % 2], dtype=torch.long)
            
            batch = {
                'input': torch.randn(1, 4),
                'label': label
            }
            yield batch
    
    def __len__(self):
        return self.num_samples

model = SimpleModel()
loader = SimpleDataset(num_samples=2, task='TASK1')
epoch = 0
task = 'TASK1'

ARGS = {
    'model': model,
    'loader': loader,
    'epoch': epoch,
    'task': task
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
sklearn, torch, tqdm — 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
majority_vote Not yet run
unverified (harvested, no recorded run)
whr000001/dell/ensemble/ensemble_utils.py
code served (permissive licence) · get_code("0df62fff81a45786")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 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 7 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

Large language models are limited by challenges in factuality and hallucinations to be directly employed off-the-shelf for judging the veracity of news articles, where factual accuracy is paramount. In this work, we propose DELL that identifies three key stages in misinformation detection where LLMs could be incorporated as part of the pipeline: 1) LLMs could \emph{generate news reactions} to represent diverse perspectives and simulate user-news interaction networks; 2) LLMs could \emph{generate explanations} for proxy tasks (e.g., sentiment, stance) to enrich the contexts of news articles and produce experts specializing in various aspects of news understanding; 3) LLMs could \emph{merge task-specific experts} and provide an overall prediction by incorporating the predictions and confidence scores of varying experts. Extensive experiments on seven datasets with three LLMs demonstrate that DELL outperforms state-of-the-art baselines by up to 16.8\% in macro f1-score. Further analysis reveals that the generated reactions and explanations are greatly helpful in misinformation detection, while our proposed LLM-guided expert merging helps produce better-calibrated predictions.

For agents

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

get_harvested_code_for_paper("2402.10426")
get_code_for_paper("2402.10426")
have("2402.10426")

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

curl https://syntology.ai/api/ran/2402.10426.json

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

[![Syntology run record](https://syntology.ai/api/ran/2402.10426.svg)](https://syntology.ai/paper/2402.10426)

Connect an agent — have() is free.