RADAR: Retrieval-Augmented Detector with Adversarial Refinement for Robust Fake News Detection

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Ma, Song-Duo, Liu, Yi-Hung, Lin, Hsin-Yu, Chen, Pin-Yu, Huang, Hong-Yan, Hsu, Shau-Yung, Chen, Yun-Nung
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914406663716864
author Ma, Song-Duo
Liu, Yi-Hung
Lin, Hsin-Yu
Chen, Pin-Yu
Huang, Hong-Yan
Hsu, Shau-Yung
Chen, Yun-Nung
author_facet Ma, Song-Duo
Liu, Yi-Hung
Lin, Hsin-Yu
Chen, Pin-Yu
Huang, Hong-Yan
Hsu, Shau-Yung
Chen, Yun-Nung
contents To efficiently combat the spread of LLM-generated misinformation, we present RADAR, a Retrieval-Augmented Detector with Adversarial Refinement for robust fake news detection. Our approach employs a generator that rewrites real articles with factual perturbations, paired with a lightweight detector that verifies claims using dense passage retrieval. To enable effective co-evolution, we introduce verbal adversarial feedback (VAF). Rather than relying on scalar rewards, VAF issues structured natural-language critiques; these guide the generator toward more sophisticated evasion attempts, compelling the detector to adapt and improve. On a fake news detection benchmark, RADAR consistently outperforms strong retrieval-augmented trainable baselines, as well as general-purpose LLMs with retrieval. Further analysis shows that detector-side retrieval yields the largest gains, while VAF and few-shot demonstrations provide complementary benefits. RADAR also transfers better to fake news generated by an unseen external attacker, indicating improved robustness beyond the co-evolved training setting.
format Preprint
id arxiv_https___arxiv_org_abs_2601_03981
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle RADAR: Retrieval-Augmented Detector with Adversarial Refinement for Robust Fake News Detection
Ma, Song-Duo
Liu, Yi-Hung
Lin, Hsin-Yu
Chen, Pin-Yu
Huang, Hong-Yan
Hsu, Shau-Yung
Chen, Yun-Nung
Computation and Language
To efficiently combat the spread of LLM-generated misinformation, we present RADAR, a Retrieval-Augmented Detector with Adversarial Refinement for robust fake news detection. Our approach employs a generator that rewrites real articles with factual perturbations, paired with a lightweight detector that verifies claims using dense passage retrieval. To enable effective co-evolution, we introduce verbal adversarial feedback (VAF). Rather than relying on scalar rewards, VAF issues structured natural-language critiques; these guide the generator toward more sophisticated evasion attempts, compelling the detector to adapt and improve. On a fake news detection benchmark, RADAR consistently outperforms strong retrieval-augmented trainable baselines, as well as general-purpose LLMs with retrieval. Further analysis shows that detector-side retrieval yields the largest gains, while VAF and few-shot demonstrations provide complementary benefits. RADAR also transfers better to fake news generated by an unseen external attacker, indicating improved robustness beyond the co-evolved training setting.
title RADAR: Retrieval-Augmented Detector with Adversarial Refinement for Robust Fake News Detection
topic Computation and Language
url https://arxiv.org/abs/2601.03981