Fake News Detection with Retrieval Augmented Generative Artificial Intelligence
Bibliographic record
Abstract
The rapid spread of false information on social media has grown to be a serious problem that influences public opinion and decision-making. Fake news spreads rapidly and extensively, often outpacing efforts to debunk or mitigate its effects. Traditional methods for detecting fake news face numerous challenges, including the necessity for extensive model training and the potential for inherent biases. Although Large Language Models (LLMs) have seen substantial improvements recently, their use in fake news detection poses the risk of producing false or misleading information due to their possible hallucinations. This study presents a new strategy to combat fake news by integrating Mixtral-8x7B, a Sparse Mixture of Experts (SMoE) Large Language Model, with a Retrieval-Augmented Generation (RAG) framework. Our framework employs Google’s search API to retrieve relevant articles in real time, harnessing Mixtral’s sophisticated language processing capabilities and RAG’s ability to access current information dynamically. Initial results are promising, indicating that our approach performs comparably to established fake news detection techniques. Our method operates without the need for extensive model training, offering significant cost savings and contributing to developing more efficient tools for detecting misinformation in the digital era, which will help stop the spread of misleading data more efficiently.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".