Conspiracy theories are anti-authoritarian narratives that lead to social conflict, impacting how people perceive political information. To help in understanding this issue, we introduce the Conspiracy Frame: a fine-grained semantic representation of conspiratorial narratives derived from frame-semantics and semiotics, which spawned the Conspiracy Frames (Con.Fra.) dataset: a corpus of Telegram messages annotated at span-level. The Conspiracy Frame and Con.Fra. dataset contribute to the implementation of a more generalizable understanding and recognition of conspiracy theories. We observe the ability of LLMs to recognize this phenomenon in-domain and out-of-domain, investigating the role that frames may have in supporting this task. Results show that, while the injection of frames in an in-context approach does not lead to clear increase of performance, it has potential; the mapping of annotated spans with FrameNet shows abstract semantic patterns (e.g., Kinship', Ingest_substance') that potentially pave the way for a more semantically- and semiotically-aware detection of conspiratorial narratives.
Figures & tables
Figure 1: Performance Analysis: Model Size and Prompt Strategy Effects
Figure 2: Span-Level F1 Score Comparison: All 70B Models
Figure 3: The ELO scores of the Few-Shot and Frame-Guided settings. Figure on the left shows the comparison between of the two approaches in CT recognition; Figure on the right on span detection.