MétaCan
Menu
Back to cohort
Record W7155652366 · doi:10.1145/3787279.3787298

TALME: Topical Adversarial LLM-based Misinformation Enforcer

2025· article· W7155652366 on OpenAlexaff
Xinyu Hu, Hanbo Yu, Zhiwei Fu, Steven H. H. Ding, Benjamin C. M. Fung, Devin Pereira, Jun Meng

Bibliographic record

Venuenot available
Typearticle
Language
FieldSocial Sciences
TopicMisinformation and Its Impacts
Canadian institutionsQueen's UniversityMcGill University
Fundersnot available
KeywordsMisinformationRumorAdversarial systemDisinformationLimitingSoftware deploymentTrustworthinessHindsight bias

Abstract

fetched live from OpenAlex

Rumor detection plays an essential role in limiting the spread of misinformation on social media, especially during fast-moving events such as natural disasters and public health crises. With the growing volume and speed of information dissemination online, the ability to accurately and robustly identify false or misleading content has become increasingly important. While recent advances in deep learning and large language models have improved automated rumor detection, many existing models still suffer from topical bias: they perform well on familiar topics but often fail to generalize to new or unseen ones, which are common in real-world scenarios. This challenge limits the deployment of rumor detection systems in dynamic environments, where new topics constantly emerge. To address this issue, we propose the Topical Adversarial Large-Language-Model-based Misinformation Enforcer (TALME), a novel framework designed to enhance cross-topic generalization. TALME combines large language models with adversarial training to reduce reliance on topic-sensitive cues and instead emphasize more generalizable linguistic signals such as writing style. Experiments on real-world rumor detection datasets demonstrate that TALME consistently outperforms strong baselines in cross-topic settings. These results highlight the importance of topic-debiasing strategies and support the development of more robust, adaptable, and trustworthy misinformation detection systems.

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 imitation

Not 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.

metaresearch head score (Codex)0.003
metaresearch head score (Gemma)0.013
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.003
Threshold uncertainty score0.016

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0030.013
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0010.000
Science and technology studies0.0010.001
Scholarly communication0.0010.002
Open science0.0020.003
Research integrity0.0020.003
Insufficient payload (model declined to judge)0.0020.001

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.

Opus teacher head0.015
GPT teacher head0.324
Teacher spread0.310 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreMethods

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".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same topicMisinformation and Its ImpactsFrench-language works237,207