SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2605.02712 · 2026

mdok-style at SemEval-2026 Task 10: Finetuning LLMs for Conspiracy Detection

Dominik Macko

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 2 functions out of this paper's own repositories and ran 2 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
kinit-sk/mdok-style-psycomark2026 canonical 2 of 2
FunctionStatusWhere it lives
anonymize Ran kinit-sk/mdok-style-psycomark2026/mdok-style.py
pointer only (licence: GPL-3.0) · get_code("7c52f16ec20ace78")
load_and_filter_data Ran kinit-sk/mdok-style-psycomark2026/mdok-style.py
pointer only (licence: GPL-3.0) · get_code("2c36dea83d855974")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

SemEval-2026 Task 10 is focused on conspiracy detection. Specifically, the goal is to detect whether a Reddit comment expresses a conspiracy belief. Our submitted mdok-style system utilizes data augmentation and self-training (to cope with a rather small amount of training data) to finetune the Qwen3-32B model for a binary text-classification task. The submitted system is very competitive, ranking in the 85th percentile (8th out of 52 submissions). The results shown that our approach, which originated in machine-generated text detection, can be used for conspiracy detection as well.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2605.02712")
get_code_for_paper("2605.02712")
have("2605.02712")

Connect an agent — have() is free.