Media bias is an enduring feature of news dissemination, reflecting the subjective perspectives of its creators across history. From archaic records like "The Victory Stele of Naram-Sin" to contemporary news channels, bias permeates media, influencing political, social, and public health narratives. This research aims to investigate the persistent phenomenon of media bias and the potential of large language models (LLMs)(Kojima et al., 2022) in its detection and classification, in order to deploy publicly available software tools aiming to enhance media literacy among news consumers.

Traditionally, media bias served the interests of ruling powers; even with the rise of modern journalism, objectivity is often compromised by commercial pressures and inherent human biases. (Rodrigo-Ginés et al., 2024). As media landscapes evolve, bias continues to shape public opinion, impacting democratic processes and public health perceptions—evident during the COVID-19 pandemic, where polarized media narratives swayed public health decisions and fueled misinformation. (Recio-Román et al., 2023)

Current research on the effects of labeling media bias or propaganda, whether automatically or with human involvement, highlights the complexity of the issue. Depending on different circumstances, labeling can lead to negative outcomes (such as reinforcing filter bubbles by providing means to avoid news with a different perspective), no change in news consumption behavior at all, or, in some cases, an actual improvement in media literacy as intended (Zavolokina et al., 2024).

This research aims to develop a technical solution for the automatic labeling of biased media content, emphasizing several proposals that we hope will lead to a positive effect on media literacy among those presented with the system’s assessments.

These proposals include using a fine-grained taxonomy of bias types rather than a simple binary left/right labeling, focusing on detailed explanations for each model decision in natural language, marking bias at the sentence level rather than at the article or publication level to provide more insights, fine-tuning autoregressive models like GPT-3.5 or Mistral with high-quality examples instead of using “simple” bidirectional models like BERT(Brown et al., 2020) or non-finetuned models, and focusing on the German language, which has not yet been properly explored for such systems.

Understanding readers' perceptions when exposed to bias-labeled content is another facet of this research. It will explore how bias labeling influences readers' views on credibility and neutrality and whether real-time bias indicators affect news consumption behaviors. As mentioned, practical applications serve as a cornerstone of this research. One aim is to implement bias detection systems in real-world settings, such as search engines and news aggregators, to promote balanced information consumption. The development of user tools, like browser extensions highlighting media bias, intends to address public need for transparent information evaluation.

In essence, this research contributes to media literacy enhancement by demystifying media bias through advanced computational methods. By refining detection mechanisms, classifying bias more effectively, and implementing practical tools, it aims to fortify democratic discourse and public understanding, thereby addressing the pervasive influence of media bias in today’s interconnected world.

References

Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., Agarwal, S., Herbert-Voss, A., Krueger, G., Henighan, T., Child, R., Ramesh, A., Ziegler, D. M., Wu, J., Winter, C., … Amodei, D. (2020). Language Models are Few-Shot Learners. In H. Larochelle, M. Ranzato, R. Hadsell, M. F. Balcan, & H. Lin (Eds.), Advances in Neural Information Processing Systems (Vol. 33, pp. 1877–1901). Curran.

Kojima, T., Gu, S. S., Reid, M., Matsuo, Y., & Iwasawa, Y. (2022). Large language models are zero-shot reasoners. Advances in Neural Information Processing Systems, 35, 22199–22213.

Recio-Román, A., Recio-Menéndez, M., & Román-González, M. V. (2023). Influence of Media Information Sources on Vaccine Uptake: The Full and Inconsistent Mediating Role of Vaccine Hesitancy. Computation (Basel). https://doi.org/10.3390/computation11100208

Rodrigo-Ginés, F.-J., Carrillo-de-Albornoz, J., & Plaza, L. (2024). A systematic review on media bias detection: What is media bias, how it is expressed, and how to detect it. Expert Systems with Applications, 237, 121641.

Keywords: Media Bias, Large Language Models, Bias Detection, Natural Language Processing, Journalism, Public Opinion, Taxonomy

Zusammenfassung

Geführte Akustische Wellen (GAW) haben sich im Themengebiet des Structural Health Monitoring (SHM) etabliert. Neben ihren Vorteilen bei der Überwachung von Objekten und Detektion von Fehlstellen, gibt es jedoch auch einige Herausforderungen. Zu diesen zählt die dispersive Natur der häufig eingesetzten Lambwellen. Dispersion führt zu Signalverzerrung und reduziert dadurch die räumliche Auflösung und erschwert die Erkennung von schwach reflektierenden Fehlstellen. In diesem Beitrag wird der Einsatz eines Phased-Array-Systems zur Delaminationserkennung bei einem Mehrschichtsystem demonstriert, bei dem dispersive Lambwellen zum Einsatz kommen. Durch das Kompensieren der Dispersionseffekte kann die Sign Coherence Factor (SCF) Erweiterung des Total Focusing Method (TFM) Algorithmus eingesetzt und so auch schwach reflektierende Fehlstellen erkannt werden. Des Weiteren wird auf das Entstehen von Modenüberlagerungen bei Mehrschichtsystemen eingegangen, die bei der Auswahl der Arbeitsfrequenz und Sendesignallänge des Phased-Array-Systems berücksichtigt werden müssen.

Guided Acoustic Waves (GAW) are well established in the field of Structural Health Monitoring (SHM). However, in addition to their advantages in monitoring objects and detecting defects, there are also several challenges. These include the dispersive nature of the commonly used Lamb waves. Dispersion leads to signal distortion that reduces spatial resolution and makes it difficult to detect weakly reflecting defects. This paper demonstrates the use of a phased array system for delamination detection in a multilayer system using dispersive Lamb waves. By compensating for the dispersion effects, the Sign Coherence Factor (SCF) extension of the Total Focusing Method (TFM) algorithm can be used to detect even weakly reflective defects. Furthermore, the occurrence of mode superposition in multilayer systems is discussed, which must be taken into account when selecting the operating frequency and transmit signal length of the phased array system.

Post-COVID syndrome affects at least 10% of individuals recovering from COVID-19. Currently, there is no causal treatment. This retrospective cohort study aimed to evaluate the potential of traditional Chinese medicine (TCM) in treating post-COVID symptoms. TCM physicians in Germany and Austria completed online questionnaires to retrospectively record symptoms, treatment approaches, and outcomes for patients diagnosed with post-COVID. Nine physicians collected data from 79 patients (65% female, 47 ± 16 SD). The most common TCM treatments for post-COVID were acupuncture (n = 66; 85%), Chinese pharmacological therapy (n = 61; 77%), and Chinese dietary counseling (n = 32; 41%). After an average of 7 ± 4 TCM consultations, physicians rated global symptom improvement as 62% ± 29%. Significant alleviation from the start of TCM treatment was observed in major symptoms, such as fatigue (P < .001), impaired physical performance (P < .001), and exertional dyspnea (P < .001). TCM treatment was associated with significant improvements in post-COVID symptoms, warranting further evaluation through randomized controlled studies.

Estimating the magnitude of placebo responses across pharmacological and nonpharmacological trials is important for understanding their influence on trial outcomes. Yet, the extent to which more intense placebo interventions like sham acupuncture yield larger analgesic responses than placebo pills, and the factors predicting these responses, remain unclear. This meta-analysis investigated the magnitude and predictors of placebo analgesia responses in pharmacological vs acupuncture trials. Analyses included individual patient data from the placebo arm of 11 randomized controlled trials (RCTs): 9 pharmacological RCTs using placebo pills (N = 2021) and 2 acupuncture RCTs using sham acupuncture (N = 747). All trials were conducted in patients with chronic nociceptive pain (osteoarthritis, N = 2068; low back pain, N = 700). The placebo response was calculated as the change in pain intensity (0-100) between baseline and week 12. A random effects model demonstrated that placebo pills and patients with osteoarthritis exhibited smaller placebo responses than sham acupuncture and patients with low back pain (both P < 0.001, marginal effects). A mixed effects model showed that route of administration interacted significantly with baseline pain, premature termination, and the presence of adverse events. Together, predictors explained 20% to 25% of the individual variance in placebo responses, whereas 75% to 80% remained unaccounted for. In summary, sham acupuncture accounted for slightly larger placebo responses than placebo pills. Since basic trial and patient parameters explained only a small portion of this variability, we might need to start considering the patient's perception of the treatment—including cognition and emotions—to better predict placebo analgesia responses.

The textual realm and the geographic/spatial realm intersect when we use human language to talk about geographic space. Various terms have been used to talk about this intersection (“geoparsing”, “georeferencing”, “toponym resolution”, “spatial grounding” etc.) and related applications such as geographic information retrieval. In this keynote, I will review some things that the community has accomplished since 2003, what occupies people’s minds at the moment, and I will raise a few research questions that would be interesting to answer, or that would unlock the potential for new kinds of applications. I conclude with some personal conjectures about how one version of the future might look like.

The increasing consumption of news online in the 21st century coincided with increased publication of disinformation, biased reporting, hate speech and other unwanted Web content.

We describe BiasScanner, an application that aims to strengthen democracy by supporting news consumers with scrutinizing news articles they are reading online. BiasScanner contains a server-side pre-trained large language model to identify biased sentences of news articles and a front-end Web browser plug-in. BiasScanner can identify and classify more than two dozen types of media bias at the sentence level, making it the most fine-grained model and only automatic application deployed as a browser plug-in. One special feature is the high-quality, LLM-generated explanations of the model’s decisions.

While prior research has addressed news bias detection, we are not aware of any automatic work that resulted in a deployed browser plug-in (c.f. also biasscanner.org for a Web demo).

The textual realm and the geographic/spatial realm intersect when we use human language to talk about geographic space. Various terms have been used to talk about this intersection (geoparsing, georeferencing, toponym resolution, spatial grounding etc.) and related applications such as geographic information retrieval. In this keynote, I will review some things that the community has accomplished since 2003, what occupies people’s minds at the moment, and I will raise a few research questions that would be interesting to answer, or that would unlock the potential for new kinds of applications. I conclude with some personal conjectures about how one version of the future might look like.