Roy Yitzchak, Noa Lavie, Ella Rabinovichcs.CL cs.SI
Political polarization has become a defining feature of online discourse, yet its long-term evolution remains poorly understood. We present a longitudinal analysis of ideological polarization in Reddit discussions by measuring semantic differences in the language used by opposing political communities. We construct temporally aligned community-specific word embeddings and quantify ideological polarization as the semantic divergence of political concepts over time. Our analysis shows that ideological polarization has increased substantially during the study period, both at the concept- and topic-level. Unlike prior computational work, which has largely focused on cross-sectional analyses or affective dimensions of polarization at a single point at time, our approach captures the evolution of ideological differences in semantic framing. The proposed framework provides a scalable method for studying the temporal dynamics of ideological polarization in large-scale social media discourse.
Vision-language models can exhibit visual concept-conditioned divergence: given images containing demographic features, corporate logos, or ideological symbols, some models produce unusually uniform responses that differ from what peer models say about the same input. These behaviors evade text-only audits because visual concepts cannot be isolated or substituted the way text tokens can. We present VISTA (Visual Inconsistency Screening Through Analysis), a black-box cross-model audit that couples semantic entropy with distribution-based divergence to flag model-specific anomalies. In a controlled study, we implant concept-conditioned stances in three VLMs via fine-tuning on small biased datasets and confirm that VISTA detects them. Auditing six VLMs across 19 topics, VISTA surfaces 142 high-suspicion cases (1.2%) and identifies selective refusal as a previously unreported divergence pattern, where models refuse demographic queries at rates varying from 0 to 65% across groups.