EU DisinfoTest: a Benchmark for Evaluating Language Models’ Ability to Detect Disinformation Narratives (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Disinformation narratives can be deceptive and disinformative, designed to sow division, distrust, and fear. |
| Approach: | They propose to evaluate the efficacy of Language Models in identifying disinformation narratives using a Human-in-the-Loop methodology. |
| Outcome: | The EU DisinfoTest evaluates language models on their ability to perform zero-shot classification of disinformation narratives versus credible narratives. |
Similar Papers
DiNaM: Disinformation Narrative Mining with Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Disinformation is a powerful force in digital media, posing serious threats such as physical harm and the erosion of democracy. |
| Approach: | They propose to use a multi-step approach to uncover disinformation narratives by using Large Language Models to detect false information and then using clustering techniques to identify underlying disinformation stories. |
| Outcome: | The proposed algorithm outperforms general-purpose narrative mining methods by 16.4–24.7%. |
Narratives at Conflict: Computational Analysis of News Framing in Multilingual Disinformation Campaigns (2024.acl-srw)
Copied to clipboard
| Challenge: | Existing methods for multilingual framing differ from those used in English-speaking world . framers often use loaded vocabularies to create political images or favor a particular point of view . |
| Approach: | They use eight years of Russian-backed disinformation campaigns to examine framing . they find that disinformation campaign consistently favors specific framers . |
| Outcome: | The proposed method underperforms and shows high disagreements in Russian-language articles . the proposed method is based on eight years of Russian-backed disinformation campaigns . |
MisinfoBench: A Multi-Dimensional Benchmark for Evaluating LLMs’ Resilience to Misinformation (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing benchmarks assess factual accuracy in isolated queries but fail to evaluate LLMs’ resilience to misinformation in interactive settings. |
| Approach: | MisinfoBench is a benchmark designed to assess LLMs’ ability to discern, resist, and reject misinformation. |
| Outcome: | MisinfoBench assesses large language models’ ability to discern, resist, and reject misinformation in interactive settings. |
Investigating Bias in LLM-Based Bias Detection: Disparities between LLMs and Human Perception (2025.coling-main)
Copied to clipboard
| Challenge: | Detecting media bias is critical due to the spread of misinformation and disinformation on social media platforms. |
| Approach: | They investigate the presence and nature of bias within large language models and its consequential impact on media bias detection. |
| Outcome: | The proposed debiasing strategies include prompt engineering and model fine-tuning. |
A LLM-based Ranking Method for the Evaluation of Automatic Counter-Narrative Generation (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for evaluating CNs are expensive, time-consuming, and subjective, but lack a universal truth and the lack of a 'universal truth' . |
| Approach: | They propose a model ranking pipeline based on pairwise comparisons of generated CNs from different models organized in a tournament-style format to improve the evaluation process. |
| Outcome: | The proposed method achieves a high correlation with human preference, with a score of 0.88, and compares chat, instruct, and base models, exploring their strengths and limitations. |
Profiling News Media for Factuality and Bias Using LLMs and the Fact-Checking Methodology of Human Experts (2025.findings-acl)
Copied to clipboard
| Challenge: | Important efforts to characterize news media outlets in terms of their political bias and factuality are labor-intensive and prone to human biases. |
| Approach: | They propose a method that emulates criteria used by professional fact-checkers to assess the factuality and political bias of an entire outlet. |
| Outcome: | The proposed method improves on baselines and with multiple LLMs. |
Quantifying Generative Media Bias with a Corpus of Real-world and Generated News Articles (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing studies focus on LLMs undertaking political questionnaires, which offers only limited insights into their biases and operational nuances. |
| Approach: | They propose to use a curated dataset to generate 56,700 synthetic articles using nine LLMs. |
| Outcome: | The proposed model can detect political biases using supervised models and LLMs. |
Attacking Misinformation Detection Using Adversarial Examples Generated by Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models can be used to attack content filtering algorithms in social media platforms. |
| Approach: | They propose to generate adversarial examples to test the robustness of social media content filtering algorithms. |
| Outcome: | The proposed model outperforms existing models in the case of propaganda, false claims, rumours and hyperpartisan news. |
How does Misinformation Affect Large Language Model Behaviors and Preferences? (2025.acl-long)
Copied to clipboard
| Challenge: | Existing studies have explored the role of Large Language Models in combating misinformation, but there is still a lack of detailed analysis on the specific aspects and extent to which LLMs are influenced by misinformation. |
| Approach: | They propose to use a benchmark to evaluate LLMs' behavior and knowledge preference toward misinformation to identify their models. |
| Outcome: | The proposed approach is based on 10,346,712 pieces of misinformation and examines knowledge conflicts and stylistic variations. |
Measuring and Benchmarking Large Language Models’ Capabilities to Generate Persuasive Language (2025.naacl-long)
Copied to clipboard
| Challenge: | Recent studies have focused on specific domains or types of persuasion, but a general study has focused on how LLMs produce persuasive text. |
| Approach: | They construct a dataset to measure and benchmark the ability of Large Language Models (LLMs) to produce persuasive text. |
| Outcome: | The proposed model can be used to generate persuasive text across domains and domains. |