=Paper=
{{Paper
|id=Vol-3782/paper4
|storemode=property
|title=On the Categorization of Corporate Multimodal
Disinformation with Large Language Models
|pdfUrl=https://ceur-ws.org/Vol-3782/paper4.pdf
|volume=Vol-3782
|authors=Ana-Maria Bucur,Sónia Gonçalves,Paolo Rosso
|dblpUrl=https://dblp.org/rec/conf/codai2/BucurGR24
}}
==On the Categorization of Corporate Multimodal
Disinformation with Large Language Models==
https://ceur-ws.org/Vol-3782/paper4.pdf
On the Categorization of Corporate Multimodal
Disinformation with Large Language Models
Ana-Maria Bucur1,2,* , Sónia Gonçalves3 and Paolo Rosso2,4
1
Interdisciplinary School of Doctoral Studies, University of Bucharest, Romania
2
PRHLT Research Center, Universitat Politècnica de València, Spain
3
Universidad de Sevilla, Spain
4
ValgrAI Valencian Graduate School and Research Network of Artificial Intelligence, Spain
Abstract
Disinformation is becoming more prevalent in the corporate sphere, especially as brands choose to promote their
products through influencers or micro-celebrities who are perceived as reliable and impartial, but may facilitate
false information. The spread of disinformation can have negative economic impacts on companies and brands,
which can even affect their reputation. Artificial Intelligence can help detect false information and has become
increasingly important in combating disinformation. The current work addresses the problem of characterizing
multimodal disinformation targeting corporations and provides a collection of content that spreads disinformation
in digital media. The content was manually annotated with information about the target (Organization, Brand,
or Other) and the source (Corporate, Advertising, or Other) of the false content. We conduct comprehensive
experiments to evaluate the effectiveness of state-of-the-art Unimodal and Multimodal Large Language Models in
identifying the source and target of the content.
Keywords
Corporate Multimodal Disinformation, Multimodal Large Language Models, Spanish
1. Introduction and Related Work
According to [1], the concept of disinformation refers to a deliberate and organized attempt to confuse
or manipulate people by providing dishonest information. In the corporate sphere, disinformation is
gaining more ground. It is orchestrated to persuade audiences and hold great appeal for advertisers
who promote their dissemination as a lure “because it fits more easily into people’s prejudices” [2].
The issue can become even more dangerous when we consider that more and more brands choose to
promote their products through influencers or micro-celebrities, which can facilitate false information
[3]. These opinion leaders are perceived with high levels of reliability and impartiality, allowing them
to recommend products and services on various social media platforms and generate word of mouth
that brands leverage for their commercialization [4].
The spread of disinformation can be a risk to companies and brands and cause a negative economic
impact [5] that can even affect their reputation. Disinformation that can impact a company’s reputation
may stem from political, financial, emotional, or internal motivations, such as discontented employees
[6]. Therefore, it is important for organizations to manage trusting relationships with the public.
Organizations can become victims of individuals and advanced technologies with the intention to
damage their reputation for twisted purposes [7] through the use of deepfakes, a new form of fake
news that threatens companies, organizations, and brands [8, 9, 10]. As the reputation of organizations
can be affected by the spread of disinformation, to protect the corporate image, communication officers
need to be aware of strategies to combat it, such as fact-checking. Artificial Intelligence has enabled the
implementation of automated approaches capable of detecting false information [11, 12], also from a
multimodal perspective [13, 14, 15, 16, 17, 18].
Proceedings of the 1st Workshop on COuntering Disinformation with Artificial Intelligence (CODAI), co-located with the 27th
European Conference on Artificial Intelligence (ECAI), pages 29–39, October 20, 2024, Santiago de Compostela, Spain
*
Corresponding author.
$ ana-maria.bucur@drd.unibuc.ro (A. Bucur); songomgon2@alum.us.es (S. Gonçalves); prosso@dsic.upv.es (P. Rosso)
0000-0003-2433-8877 (A. Bucur); 0000-0002-5579-7761 (S. Gonçalves); 0000-0002-8922-1242 (P. Rosso)
© 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
ceur-ws.org
Workshop ISSN 1613-0073
Proceedings
29
Ana-Maria Bucur et al. CODAI Workshop Proceedings 29–39
Figure 1: Selected examples of false content. The data is diverse, containing screenshots from social
media, websites, etc. Translated text, first image: “Results for Chueca”. Translated text, second
image: “Get Dyson V11 for only 1,95 euros. Fill in the short questionnaire and respond to the
three questions...”. Translated text, third image: “Congratulations! Repsol 35th anniversary
government subsidy! Through the questionnaire, you will have the opportunity to obtain 1000
euros.”. Translated text, fourth image: “Bad news for the climate fanatics: with 661 gigatons of
extra mass, Antarctica continues to expand...”.
Unlike general disinformation, which can target individuals, events, or broad societal issues, corporate
disinformation often has direct financial implications and can damage trust in brands and organizations.
Recognizing the unique characteristics and potential impacts of such disinformation, our work aims to
deepen the understanding of what are the actors targeted by corporate disinformation and the sources
spreading it. By classifying the target of the false content, we can identify whether the affected entity is
an organization or a brand. Furthermore, identifying the source will enable affected entities to take
action and develop appropriate responses to counter the disinformation being spread about them.
As there are many previous works on multimodal fake content detection [18, 14, 13, 16, 17], we aim
to characterize content that has been already fact-checked and confirmed as false. To the best of our
knowledge, this is the first time that the problem of multimodal disinformation targeting corporations
has been addressed automatically. For this purpose, a collection of multimodal content in Spanish
that was already fact-checked is collected and annotated by expert annotators with information about
the target and source of the content (Figure 1). Our dataset consists of 534 samples, together with
annotations for the target (Organization, Brand, or Other) and the source (Corporate, Advertising,
or Other) spreading disinformation. The false content can be targeted at an Organization, such as
a company, institution, or an individual representing them. It can also target a Brand or a person
associated with it. Alternatively, disinformation can be classified as Other, meaning it is not aimed at an
organization or brand but contains misleading information intended to deceive the general population.
Furthermore, false content can originate from various sources. It may stem from a Corporate origin,
where a corporate entity is responsible for spreading disinformation, rather than just an individual.
Alternatively, it could be a result of persuasive Advertising, typically in the form of paid posts on social
media. Lastly, false content may originate from Other sources, such as online users disseminating
misleading information.
In this paper, we address the problem of characterizing multimodal disinformation targeting corpora-
tions. Our work makes the following contributions:
• A collection of multimodal false content (visual and textual information in Spanish) that spread
disinformation in digital media on corporations is compiled and annotated with information
about the source and target of the false content;
• Comprehensive experiments are conducted to evaluate the effectiveness of state-of-the-art Uni-
modal and Multimodal Large Language Models (LLMs) in characterizing false content.
30
Ana-Maria Bucur et al. CODAI Workshop Proceedings 29–39
2. Data Collection
The dataset used in this work is obtained from the IBERIFIER repository1 , which includes online content
that has been fact-checked and verified2 . IBERIFIER is a project that aims to fight disinformation in
digital media in Spain and Portugal, in which data from various fact-checking websites is collected
and analyzed. In our research, we specifically focus on false content in Spanish that was verified by
EFE Verifica3 and Maldita.es4 , as these organizations contributed the most content to the IBERIFIER
database. Our dataset consists solely of posts that were confirmed by these fact-checking entities to
contain false information. This limits the dataset size, as obtaining fact-checked data is challenging. Our
dataset contains 496 samples from Maldita.es and 38 samples from EFE Verifica, with multimodal data
represented through both visual and textual information in Spanish. By deliberately focusing on posts
that have been verified to contain disinformation, we can more effectively evaluate the performance of
pre-trained visual transformer models and LLMs in characterizing deceptive information. This dataset
allows us to study and understand how these models identify the different targets and sources spreading
disinformation. The dataset is an essential resource for studying the effectiveness of LLMs in classifying
false content from visual and textual cues found in images.
Figure 2: The format of the false content found Figure 3: Platforms used to spread the false con-
in the collected data: pictures, screen- tent. Most of the content was shared on
shots from social media platforms, from social media platforms and WhatsApp.
different websites, or news articles.
For each of the collected images, we also retrieved information about the format of the content and
the platform used to spread it using the IBERIFIER API. In Figure 2, we present the various formats
of false content. The most common type of false content is represented by pictures, followed by
screenshots from social media. Figure 3 shows the platforms used to spread the disinformation content.
The data suggests that social media platforms like Twitter, Facebook, TikTok, and Instagram are the
primary channels used to spread false content. However, we found that a considerable amount of false
information is also shared through messaging apps like WhatsApp.
Two expert annotators have labeled each instance of false content with information about the target
and source. The target of the disinformation can be an Organization (either a company, an institution,
or a person representing it), a Brand (or a person representing it), or it can be Other, meaning that it is
not targeted towards an organization or a brand, and it contains false information intending to mislead
1
https://iberifier.eu/
2
https://iberifier.eu/factchecks/
3
https://verifica.efe.com/
4
https://maldita.es/
31
Ana-Maria Bucur et al. CODAI Workshop Proceedings 29–39
the general population about various topics, such as climate change, immigrants, conspiracy theories,
local news. With regard to the different sources of false content (i.e. the origin of the content), the
content can be of Corporate origin (usually, there is an entire corporate entity behind the spread of
disinformation, not just an individual), persuasive Advertising (usually paid posts on social media),
or Other - usually false content spread by other users. The Other class also contains false content in
which the identity of the spreader does not appear in or cannot be inferred from the image/text (see
Figure 1, 1st and 4th example). We obtained a strong agreement between the two annotators (Cohen’s
𝜅 0.90). The disagreements between them have been resolved by a senior researcher in the field. The
final dataset contains 347 samples targeting an organization, 87 targeting a brand, and 100 targeting
other entities. Regarding the sources of the false content, the dataset is comprised of 52 Corporate, 4
Advertising, and 478 Other sources.
We showcase 4 examples from the collected data in Figure 1. The dataset includes different types
of disinformation found in digital media, which makes it difficult to identify the source and target
spreading the content. The first example shows an image with a figure representing the electoral results
from the Chueca neighborhood of Madrid. However, the image is spreading disinformation because the
results are actually from a municipality in Toledo with the same name. This is a classic example of how
disinformation can be spread by manipulating images and providing false information. The source of
the content was classified as Other because the origin of the information is unknown, it does not appear
in the text or the image. On the other hand, the target is Organization because the disinformation
publication affects one or more organizations, in this case, political parties (People’s Party (PP)) and
Spanish Socialist Workers’ Party (PSOE)).
The second example is a sponsored post from Facebook, asking individuals to complete a brief
questionnaire for the chance to purchase a discounted vacuum cleaner. However, this image represents
a classic phishing post where individuals are persuaded to share their banking information with
malicious entities. This example illustrates how social media platforms can be used to spread phishing
scams that can deceive unsuspecting users. The source of the content was categorized as Advertising
due to the information originating from a clearly identified advertising publication (sponsored content),
indicating that the advertising is conducted on a social network through payment. Conversely, the
target is identified as Brand because the disinformation publication impacts brands, specifically Dyson
and Lidl.
The third example is a screenshot from a website that claims to be of Repsol S.A., an energy and
petrochemical company from Spain. However, the website is not the real website of the company,
and it is used for phishing. Malicious actors are using the website to trick users into sharing their
personal data. The content was categorized as Corporate because the web page appears to be created by
a corporate entity rather than an individual. On the other hand, the target is Brand, as it targets Repsol.
In the fourth example, we present a screenshot from social media that is not targeted towards a
corporate entity or a brand, and it was labeled as Other - trying to mislead the general population. The
source of the content was labeled as Other, with no information about the source provided in the text
or image.
3. Methodology
We perform experiments in zero-shot or few-shot settings to evaluate the effectiveness of state-of-the-art
visual transformer models and LLMs in characterizing false content within multimodal data.
3.1. Pre-trained Visual Transformer Models
Pre-trained visual transformer models, such as CLIP [19], have shown great performance on downstream
tasks without additional training, obtaining competitive results with a supervised baseline. CLIP was
pre-trained in a self-supervised manner on a large collection of image-text pairs with a contrastive
learning objective. The model was trained to maximize similarity between pairs of the same class and
minimize similarity between pairs of different classes. CLIP extracts embeddings by processing the
32
Ana-Maria Bucur et al. CODAI Workshop Proceedings 29–39
Figure 4: Zero-Shot Classification pipeline for state-of-the-art visual transformer models: CLIP, Open-
CLIP, MetaCLIP, SigLIP. Images and class names/descriptions are passed through frozen
encoder models, and the final prediction is represented by the text that is most similar to a
given image.
image and text through a visual and textual encoder, respectively. The embeddings are then mapped
to a shared space where similarities between image-text pairs can be computed. Pre-training allows
CLIP to represent images and text with similar content closer in the embedding space while unrelated
image-text pairs are represented further apart. In this way, the model can compute the relationship
between a given image and its corresponding textual description.
We are exploring the effectiveness of using CLIP and similar models [20, 21] for zero-shot classification.
To achieve this, we investigate how well the models can predict the target and the source of online
disinformation. The zero-shot classification pipeline is presented in Figure 4. The process involves
passing images and texts, in our case, the names/descriptions of the categories, through frozen visual
and textual encoder models. The similarity between the image and each category name/description
is computed, and the category with the highest similarity score is selected as the final prediction. We
conducted our experiments in two settings: by providing the class names as labels and by providing
a short definition/description of the content we expect to find for each class. The two types of label
names, short and long, are shown in Figure 4. For target classification, we first experimented with short
label names such as Organization, Brand, and Other. We also experimented with longer names, such as
“a screenshot of false information targeting an organization (a company or an institution)”, etc. Inspired
by recent works highlighting the importance of the definitions of the concepts [22], we added more
information to the text describing the categories. For the source classification, we followed a similar
approach and experimented with both the short label names, such as Corporate, Advertising, and Other,
and longer variants.
In our experiments, we have tested the abilities of various pre-trained transformer models like
CLIP [19], OpenCLIP [23], MetaCLIP [20], SigLIP [21]. CLIP and OpenCLIP [23] have identical vision
transformer architecture, but OpenCLIP was trained on the open-source dataset LAION-2B [24], whereas
CLIP was trained on a private dataset of image-text pairs. MetaCLIP [20] uses the same architecture
and training regime as above, but the authors ensure that only high-quality image-text pairs are used
for pre-training. SigLIP [21] replaces the softmax-based contrastive loss from CLIP with a sigmoid loss.
We experiment with different variants of the models, either base, large, or huge, if available.
3.2. Large Language Models
With the great success of leveraging LLMs in various vision and language tasks [25, 26, 27, 28], we
also choose to test their abilities in characterizing multimodal disinformation shared in digital media.
We experiment with two LLMs that have shown good results in language tasks, LLaMa-2 [27], and
Mistral [25]. LLaMa is a competitive model, with good results over a suite of benchmarks related to
commonsense reasoning, word knowledge, reading comprehension, etc. [27]. Mistral is another LLM
33
Ana-Maria Bucur et al. CODAI Workshop Proceedings 29–39
Figure 5: Zero-Shot Classification pipeline with LLaVA. LLaVa uses a language model (in our case,
LLaMa) to process both visual information and language instructions, and generate an appro-
priate response. LLaVa leverages a pre-trained CLIP model to encode visual information from
images. These embeddings are then projected into the same word embeddings space and fed
into LLaMa. Finally, LLaMa generates a suitable language response.
that surpasses LLaMa-2 on all the tested benchmarks [25]. We chose these two models to evaluate
their classification performance on our dataset based solely on the text found in the image and its
caption. The text found in images is written in Spanish (as presented in Figure 1) and was extracted
using Pytesseract5 . The caption of the image was generated using BLIP-2 [29]. We conducted zero-shot
and few-shot experiments using the aforementioned LLMs.
Although these LLMs are pre-trained on data that is mostly in English, LLaMa, for example, was
pre-trained on 1.3B Spanish tokens (0.13% of the total corpus). This amount of pre-training tokens
makes it capable of processing Spanish content, although the results may not be as accurate as for
English data [30]. No information about the data used for pre-training Mistral models is available [25].
Because the text from the multimodal false content is in Spanish, we chose to include in our experiments
a fine-tuned version of LLaMa-7B on Spanish instructions6 .
3.3. Multimodal Large Language Models
In our work, we also conduct experiments using the Multimodal LLM LLaVa [31], which is a general-
purpose visual and language model (Figure 5). LLaVa uses a language model (in our case, LLaMa-2
[27]) to process both the visual information from the image and the text of the language instructions.
LLaVa uses a pre-trained CLIP vision transformer to process visual input, which is then projected in
the same embedding space as the text. The visual and text embeddings are then fed to LLaMa, which
generates a suitable language response. In our experiments we use LLaVA-v1.5 [26] and LLaVA-v1.5
Q-Instruct [28]. We chose to use LLaVA-v1.5, as it is an improved version of the original LLaVA,
and it achieves state-of-the-art results on various benchmarks related to visual question answering.
LLaVA-v1.5 Q-Instruct improves over the aforementioned versions by demonstrating low-level visual
perception [28].
5
https://github.com/madmaze/pytesseract
6
clibrain/Llama-2-7b-ft-instruct-es
34
Ana-Maria Bucur et al. CODAI Workshop Proceedings 29–39
4. Experimental Setup
As part of our experiments, we tested the zero-shot and few-shot (one-shot) capabilities of various
models. Our test set is comprised of 519 samples, as 15 samples were kept to potentially be used for the
few-shot settings. We used the open-source implementations for all the models. Due to computational
limitations, we only experimented with 7B variants of LLMs and Multimodal LLMs. While generating
the output, we use the default temperature of 0.7. Additionally, we post-processed the generated output
to remove any punctuation, quotation marks, or explanations generated by the models. The prompts
for LLaMa-2-7B and Mistral-7B were written in English. For LLaMa-2-7B-ES, given that it is a model
fine-tuned for the Spanish language, we use prompts written in Spanish.
5. Results
Target Source
Model Labels Weighted-F1 Brand Org. Other Weighted-F1 Adv. Corp. Other
CLIP𝑏𝑎𝑠𝑒 Short 29.62 28.38 29.17 32.31 40.57 1.37 15.71 43.16
Long 47.89 29.20 57.89 27.99 84.62 6.90 48.78 88.45
CLIP𝑙𝑎𝑟𝑔𝑒 Short 32.77 25.37 36.49 25.77 43.19 1.29 6.40 43.19
Long 49.95 27.20 59.01 36.97 78.97 2.50 32.52 83.83
MetaCLIP𝑏𝑎𝑠𝑒 Short 20.80 31.81 15.10 31.80 48.73 1.22 12.32 52.49
Long 50.53 30.38 60.03 33.71 70.51 3.28 34.21 74.35
MetaCLIP𝑙𝑎𝑟𝑔𝑒 Short 19.46 45.69 8.82 35.15 83.69 2.90 26.51 89.62
Long 14.99 26.67 7.57 31.62 80.46 2.20 41.90 84.56
MetaCLIPℎ𝑢𝑔𝑒 Short 13.04 20.83 6.13 31.22 82.48 5.71 14.74 89.43
Long 54.34 28.00 66.37 33.78 85.36 8.70 40.00 90.11
OpenCLIP𝑏𝑎𝑠𝑒 Short 10.10 25.11 0.58 31.42 82.73 3.70 0.00 91.14
Long 36.64 31.10 38.10 36.18 63.83 1.93 23.66 68.02
OpenCLIP𝑙𝑎𝑟𝑔𝑒 Short 18.66 32.34 10.87 34.88 76.72 2.38 25.00 82.08
Long 23.29 36.88 17.19 33.53 33.86 1.06 30.77 34.31
OpenCLIPℎ𝑢𝑔𝑒 Short 55.05* 45.54 62.41 36.75 65.52 3.01 8.33 71.37
Long 21.42 28.32 15.83 35.57 78.21 1.69 22.78 83.95
SigLIP𝑏𝑎𝑠𝑒 Short 21.82 31.40 16.40 33.02 82.90 10.53 7.23 90.60
Long 29.54 29.43 28.02 35.10 17.14 0.96 29.17 16.03
SigLIP𝑙𝑎𝑟𝑔𝑒 Short 13.91 37.21 2.87 33.51 86.18* 0.00 9.52 94.03
Long 51.59 30.45 59.93 39.81 4.16 0.96 34.01 1.26
Table 1: Zero-shot classification using visual transformer models. We present the Weighted F1 -score,
and the F1 -scores for each of the classes. We present the best results with bold, and with
underline the second-best results. * denotes statistically significant differences between best
and second-best models using the McNemar-Bowker Test (p<0.05).
We evaluate each model for the two tasks, either target or source classification, by computing F1
scores for each class. We also measure the performance over each task using Weighted-F1 score, given
that the categories of our dataset are highly imbalanced. We present the results of the zero-shot
classification using CLIP, MetaCLIP, OpenCLIP, and SigLIP in Table 1. For the majority of the models
and variants, using longer descriptions of the class names improved the results of the classification.
The best model for classifying the target of the false multimodal content was OpenCLIPℎ𝑢𝑔𝑒 , obtaining
a Weighted-F1 score of 55.05%. Even if SigLIP𝑙𝑎𝑟𝑔𝑒 obtained an 86.18% Weighted-F1 score for predicting
the source of disinformation, it cannot accurately make predictions for all the categories.
In Table 2, we showcase the performance of the LLMs in zero-shot and few-shot settings. LLaMa-2-7B,
Mistral-7B and LLaMa-2-7B-ES use only the text extracted from the image and its generated caption.
By providing only one example in the prompt, the performance of LLaMa-2-7B improves by 28.15%.
For Mistral-7B, there is a 10.49% improvement in Weighted-F1 score for target classification, while,
for LLaMa-2-7B-ES, the improvement is minimal between zero-shot and few-shot settings. However,
the model fine-tuned on Spanish instructions, LLaMa-2-7B-ES, obtained the best Weighted F1 score of
64.01% in the few-shot setting and second-best Weighted F1 score of 62.31% in the zero-shot setting.
35
Ana-Maria Bucur et al. CODAI Workshop Proceedings 29–39
Target Source
Model Weighted-F1 Brand Org. Other Weighted-F1 Adv. Corp. Other
LLaMa-2-7B (zero-shot) 14.33 0.00 12.90 31.85 80.71 0.00 0.00 88.94
LLaMa-2-7B (one-shot) 42.48 22.43 50.47 31.00 72.66 2.65 0.00 80.05
Mistral-7B (zero-shot) 49.89 23.53 59.51 38.04 86.98 0.00 4.26 95.43
Mistral-7B (one-shot) 60.38 32.00 74.89 32.62 86.35 0.00 0.00 95.15
LLaMa-2-7B-ES (zero-shot) 62.31 19.23 76.07 50.00 81.81 2.38 41.24 86.11
LLaMa-2-7B-ES (one-shot) 64.01* 24.56 76.41 53.42 78.67 2.96 41.03 82.67
Table 2: Zero-shot and one-shot classification using LLMs. * LLaMa-2-7B-ES (one-shot) obtains statistically
significant improvement over the best English counterpart Mistral-7B (one-shot) in Target
prediction (McNemar-Bowker Test, p<0.05).
Target Source
Model Weighted-F1 Brand Org. Other Weighted-F1 Adv. Corp. Other
LLaVA-v1.5-7B 51.88* 21.37 65.85 27.89 61.68 1.89 8.60 67.12
LLaVA-v1.5-7B (Q-Instruct) 49.68 24.84 60.20 33.22 68.72* 2.65 15.93 74.16
Table 3: Zero-shot classification using LLaVA. * denotes statistically significant differences between
best and second-best models using the McNemar-Bowker Test (p<0.05).
Predicting the target of disinformation is easier, usually relying on specific cues, such as the presence
of organizations’ or brands’ logos or names appearing in the picture or written in text. However,
predicting the source of disinformation from multimodal content is a harder task, as in many instances,
no information about it appears, and the source is unknown. For source classification, the LLMs
sometimes only predict the Other class, failing to predict other categories. Using the LLaMa-2-7B-ES
in one-shot setting with the text from the image and its caption as input was proven to be a suitable
approach for target classification, surpassing all other visual models, such as CLIP, MetaCLIP, OpenCLIP
and SigLIP. The limitations of general language models trained solely on English data are highlighted by
the best performance of LLaMa-2-7B-ES, which was adapted to Spanish data. This further emphasizes
the need to develop language-specialized LLMs.
In Table 3, we show the results of LLaVA-v1.5-7B for zero-shot classification. LLaVA-v1.5-7B obtains
a better performance of 51.88% Weighted-F1 score for target classification, while LLaVA-v1.5-7B (Q-
Instruct) obtains a better performance for source classification (74.16% Weighted-F1 score). In zero-shot
settings, LLaVA-v1.5-7B outperforms the English-based language-only counterparts, LLaMa-2-7B and
Mistral-7B, for target classification, obtaining a Weighted-F1 score of 51.88%. However, it has a lower
performance than LLaMa-2-7B-ES. According to our experiments, while general LLMs pre-trained
on mostly English data can provide satisfactory results for identifying false content in our corporate
multimodal disinformation dataset, models specifically adapted for a particular language perform better.
This is because they can make use of the Spanish text present in the multimodal content, leading to
enhanced performance.
6. Conclusion
In this paper, our aim was to create a valuable resource for characterizing corporate multimodal
disinformation from digital media featuring both visual and textual elements in Spanish, annotated with
details about the source and target of the false content. By publishing our dataset, we aim to encourage
further research in this area and the development of more effective disinformation characterization
technologies. Our comprehensive experiments have assessed the efficacy of state-of-the-art multimodal
transformer models and LLMs in characterizing false content within images. Our findings reveal that
predicting the target of the false content is easier than predicting the source, as the latter requires
information that may not be easily represented in the multimodal data. In terms of zero-shot versus few-
shot settings, providing one example for each class improved the performance for target classification by
28.15% for LLaMa-2-7B and 10.49% for Mistral-7B in terms of Weighted-F1 score. LLaVA, the Multimodal
36
Ana-Maria Bucur et al. CODAI Workshop Proceedings 29–39
LLM that we had tested, obtained a Weighted-F1 score of 51.88% in a zero-shot setting for target
classification. The best result for target classification, of 64.01% Weighted-F1 score, was obtained by
LLaMa-2-7B-ES in one-shot setting, suggesting that LLMs specifically adapted for a particular language
are needed when processing non-English data.
Our goal is to assist corporate entities in monitoring digital streams for fake news that could potentially
harm their reputations. In our future work, we intend to expand our dataset and develop methods for
identifying the specific brands and organizations targeted by false content. Moreover, we would like to
expand our analysis to recently-released LLMs, such as LLama-37 , LLaVA-NeXT8 , GPT-4V [32], Gemini
Pro9 , InstructBLIP [33].
Limitations
One of the limitations of the current study is the small and imbalanced number of samples in each
class from the collected dataset. Our approach relies on data that was already fact-checked, which
is challenging to obtain. Due to the insufficient samples in some categories, our models struggle to
accurately predict those classes. To address this limitation, our future work will focus on expanding the
dataset. Specifically, we will target the collection of more samples for underrepresented classes, such as
Brand for target classification and Corporate and Advertising for source classification.
Another limitation is the use of 7B variants of LLMs and Multimodal LLMs in our experiments due
to computational limitations. Even if LLaMa-2-7B-ES and LLaVA-v1.5-7B have shown promising results
of 64.01% and 51.88% Weighted-F1 for source classification, using bigger variants of the models could
lead to further improvements in the results [34].
Acknowledgments
The work of Paolo Rosso was in the framework of FAKE news and HATE speech (FAKEnHATE-
PdC) funded by MCIN/AEI/10.13039/501100011033 and by European Union NextGenerationEU/PRTR
(PDC2022-133118-I00), Iberian Digital Media Observatory (IBERIFIER Plus) funded by the EC (DIGITAL-
2023-DEPLOY-04) under reference 101158511, and Malicious Actors Profiling and Detection in Online So-
cial Networks Through Artificial Intelligence (MARTINI) funded by MCIN/AEI/ 10.13039/501100011033
and by European Union NextGenerationEU/PRTR (PCI2022-135008-2).
References
[1] C. Ireton, J. Posetti, Journalism, fake news & disinformation: handbook for journalism education
and training, Unesco Publishing, 2018.
[2] P. Berthon, E. Treen, L. Pitt, How truthiness, fake news and post-fact endanger brands and what
to do about it, NIM Marketing Intelligence Review 10 (2018) 18–23.
[3] S. A. Baker, Alt. health influencers: how wellness culture and web culture have been weaponised
to promote conspiracy theories and far-right extremism during the covid-19 pandemic, European
Journal of Cultural Studies 25 (2022) 3–24.
[4] M. De Veirman, V. Cauberghe, L. Hudders, Marketing through instagram influencers: the impact of
number of followers and product divergence on brand attitude, International journal of advertising
36 (2017) 798–828.
[5] A. Christov, et al., Economic effects of the fake news on companies and the need of new pr
strategies, Journal of Sustainable Development 8 (2018) 41–49.
[6] A. Reid, What’s the damage?. measuring the impact of fake news on corporate reputation can act
as a guide for companies to navigate a post-truth landscape, CommunicationDirector.com (2017).
7
https://ai.meta.com/blog/meta-llama-3/
8
https://llava-vl.github.io/blog/2024-01-30-llava-next/
9
https://deepmind.google/technologies/gemini/pro/
37
Ana-Maria Bucur et al. CODAI Workshop Proceedings 29–39
[7] M. Peterson, A high-speed world with fake news: brand managers take warning, Journal of
Product & Brand Management 29 (2020) 234–245.
[8] W. A. Galston, Is seeing still believing? the deepfake challenge to truth in politics, Brookings
Institution (2020).
[9] S. Gomes-Gonçalves, Los deepfakes como una nueva forma de desinformación corporativa–una
revisión de la literatura, IROCAMM: International Review of Communication and Marketing Mix,
5 (2), 22-38. (2022).
[10] M. Westerlund, The emergence of deepfake technology: A review, Technology innovation
management review 9 (2019).
[11] M. Babakar, W. Moy, The state of automated factchecking, Full Fact 28 (2016).
[12] G. Ruffo, A. Semeraro, A. Giachanou, P. Rosso, Studying fake news spreading, polarisation
dynamics, and manipulation by bots: A tale of networks and language, Computer science review
47 (2023) 100531.
[13] Y. Li, B. Jiang, K. Shu, H. Liu, Toward a multilingual and multimodal data repository for covid-19
disinformation, in: IEEE Big Data, IEEE, 2020, pp. 4325–4330.
[14] Q. Li, M. Gao, G. Zhang, W. Zhai, J. Chen, G. Jeon, Towards multimodal disinformation detection
by vision-language knowledge interaction, Information Fusion 102 (2024) 102037.
[15] G. Zhang, A. Giachanou, P. Rosso, Scenefnd: Multimodal fake news detection by modelling scene
context information, Journal of Information Science (2022).
[16] S. Tufchi, A. Yadav, T. Ahmed, A comprehensive survey of multimodal fake news detection tech-
niques: advances, challenges, and opportunities, International Journal of Multimedia Information
Retrieval 12 (2023) 28.
[17] A. Wilson, S. Wilkes, Y. Teramoto, S. Hale, Multimodal analysis of disinformation and misinforma-
tion, Royal Society Open Science 10 (2023) 230964.
[18] A. Bondielli, P. Dell’Oglio, A. Lenci, F. Marcelloni, L. C. Passaro, M. Sabbatini, Multi-fake-detective
at evalita 2023: Overview of the multimodal fake news detection and verification task, CEUR
Workshop Proceedings (2023).
[19] A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Agarwal, G. Sastry, A. Askell, P. Mishkin,
J. Clark, et al., Learning transferable visual models from natural language supervision, in:
Proceedings of ICML, 2021, pp. 8748–8763.
[20] H. Xu, S. Xie, X. E. Tan, P.-Y. Huang, R. Howes, V. Sharma, S.-W. Li, G. Ghosh, L. Zettlemoyer,
C. Feichtenhofer, Demystifying clip data, in: Proceedings of ICLR, 2023.
[21] X. Zhai, B. Mustafa, A. Kolesnikov, L. Beyer, Sigmoid loss for language image pre-training, in:
Proceedings of ICCV, 2023.
[22] Y. Peskine, D. Korenčić, I. Grubisic, P. Papotti, R. Troncy, P. Rosso, Definitions matter: Guiding gpt
for multi-label classification, in: Findings of ACL: EMNLP 2023, 2023, pp. 4054–4063.
[23] G. Ilharco, M. Wortsman, R. Wightman, C. Gordon, N. Carlini, R. Taori, A. Dave, V. Shankar,
H. Namkoong, J. Miller, H. Hajishirzi, A. Farhadi, L. Schmidt, Openclip, 2021.
[24] C. Schuhmann, R. Beaumont, R. Vencu, C. Gordon, R. Wightman, M. Cherti, T. Coombes, A. Katta,
C. Mullis, M. Wortsman, et al., Laion-5b: An open large-scale dataset for training next generation
image-text models, in: Proceedings of NeurIPS, volume 35, 2022, pp. 25278–25294.
[25] A. Q. Jiang, A. Sablayrolles, A. Mensch, C. Bamford, D. S. Chaplot, D. d. l. Casas, F. Bressand,
G. Lengyel, G. Lample, L. Saulnier, et al., Mistral 7b, arXiv preprint arXiv:2310.06825 (2023).
[26] H. Liu, C. Li, Y. Li, Y. J. Lee, Improved baselines with visual instruction tuning, in: Proceedings of
ITIF Workshop, 2023.
[27] H. Touvron, L. Martin, K. Stone, P. Albert, A. Almahairi, Y. Babaei, N. Bashlykov, S. Batra, P. Bhar-
gava, S. Bhosale, et al., Llama 2: Open foundation and fine-tuned chat models, arXiv preprint
arXiv:2307.09288 (2023).
[28] H. Wu, Z. Zhang, E. Zhang, C. Chen, L. Liao, A. Wang, K. Xu, C. Li, J. Hou, G. Zhai, et al., Q-
instruct: Improving low-level visual abilities for multi-modality foundation models, arXiv preprint
arXiv:2311.06783 (2023).
[29] J. Li, D. Li, S. Savarese, S. Hoi, Blip-2: bootstrapping language-image pre-training with frozen
38
Ana-Maria Bucur et al. CODAI Workshop Proceedings 29–39
image encoders and large language models, in: Proceedings of ICML, 2023.
[30] H. Choi, Y. Yoon, S. Yoon, K. Park, How does fake news use a thumbnail? clip-based multimodal
detection on the unrepresentative news image, in: Proceedings of the CONSTRAINT Workshop,
2022, pp. 86–94.
[31] H. Liu, C. Li, Q. Wu, Y. J. Lee, Visual instruction tuning, in: Proceedings of NeurIPS, 2024.
[32] OpenAI, Gpt-4v(ision) system card, preprint (2023).
[33] W. Dai, J. Li, D. Li, A. M. H. Tiong, J. Zhao, W. Wang, B. Li, P. Fung, S. Hoi, Instructblip: Towards
general-purpose vision-language models with instruction tuning, 2023. arXiv:2305.06500.
[34] J. Lucas, A. Uchendu, M. Yamashita, J. Lee, S. Rohatgi, D. Lee, Fighting fire with fire: The dual
role of llms in crafting and detecting elusive disinformation, in: Proceedings of EMNLP, 2023, pp.
14279–14305.
39