Political polarization on social media intensifies during elections, when echo chambers harden partisan divisions and amplify online harms at the moment collective decisions are being made. Classical opinion dynamics models such as Friedkin-Johnsen provide a framework for studying belief change but abstract away the textual content that drives opinion shifts on social platforms. LLM-powered agent-based models reintroduce language to simulation, yet their fidelity to real political discourse—as opposed to synthetic personas or factual-consensus domains—remains untested against classical baselines. We construct a benchmark from the 2024 Indonesian Presidential Election: ~1,000 agents, 70,336 interactions, and five topics spanning four weeks of Indonesian-language discourse—to our knowledge the first opinion-dynamics benchmark grounded in non-English political data. We propose a two-axis evaluation framework pairing 1-Wasserstein distance with 3-class buffered-sign F1, reported on aggregate and per-topic tracks. Under this framework, we evaluate three open-weight LLMs (Large Language Models) alongside Friedkin Johnsen and per-agent Ridge regression, analyzing how model family, interaction topology, and sentiment polarity shape simulation fidelity.

While the field of vision-language (VL) has achieved remarkable success in integrating visual and textual information across multiple languages and domains, there is no dedicated framework for assessing human-centric alignment in vision-language systems.

We offer two contributions to address this gap. First, we introduce Anthropogenic Regional Adaptation: a regional-adaptation framework that aims to optimize model relevance to specific regional contexts while ensuring the retention of global generalization capabilities. Second, we present a simple, but effective adaptation method named Geographical-generalization-made-easy (GG-EZ), which utilizes regional data filtering and model merging.

Through comprehensive ablations on 3 VL architectures: large vision-language models (LVLM), text-to-image diffusion models, and vision-language embedding models, and a case study in Southeast Asia (SEA) regional adaptation, we demonstrate the importance of Anthropogenic Regional Adaptation and GG-EZ, showing 5-15% gains in cultural relevance metrics across SEA while maintaining over 98% of global performance and even occasionally surpassing it.

Our findings establish Anthropogenic Regional Alignment as a foundational paradigm towards applicability of multimodal VL architectures in diverse regions and demonstrate a simple-yet-effective baseline method that optimizes regional value alignment while preserving global generalization.

Online discourse is increasingly trapped in a vicious cycle where polarizing language fuelstoxicity and vice versa. Identity, one of the most divisive issues in modern politics, oftenincreases polarization. Yet, prior NLP research has mostly treated toxicity and polarization asseparate problems. In Indonesia, the world’s third-largest democracy, this dynamic threatens democratic discourse, particularly in online spaces. We argue that polarization and toxicity must be studied in relation to each other. To this end, we present a novel multi-label Indonesian dataset annotated for toxicity, polarization, and annotator demographic information. Benchmarking with BERT-base models and large language models (LLMs) reveals that polarization cues improve toxicity classification and vice versa. Including demographic context further enhances polarization classification performance.

Indonesia is rich in languages and scripts. However, most NLP progress has been made using romanized text. In this paper, we present NusaAksara, a novel public benchmark for Indonesian languages that includes their original scripts. Our benchmark covers both text and image modalities and encompasses diverse tasks such as image segmentation, OCR, transliteration, translation, and language identification. Our data is constructed by human experts through rigorous steps. NusaAksara covers 8 scripts across 7 languages, including low-resource languages not commonly seen in NLP benchmarks. Although unsupported by Unicode, the Lampung script is included in this dataset. We benchmark our data across several models, from LLMs and VLMs such as GPT-4o, Llama 3.2, and Aya 23 to task-specific systems such as PP-OCR and LangID, and show that most NLP technologies cannot handle Indonesia’s local scripts, with many achieving near-zero performance.

Online hate speech propagation is a complex issue, deeply influenced by both the perpetrator and the target’s cultural, historical, and societal contexts. Consequently, developing a universally robust hate speech classifier for diverse social media texts remains a challenging and unsolved task. The lack of mechanisms to track the spread and severity of hate speech further complicates the formulation of effective solutions. In response to this, to monitor hate speech in Indonesia during the recent 2024 presidential election, we have employed advanced Natural Language Processing (NLP) technologies to create an improved hate speech classifier tailored for a narrower subset of texts; specifically, texts that target vulnerable groups that have historically been the targets of hate speech in Indonesia. Our focus is on texts that mention these six vulnerable minority groups in Indonesia: Shia, Ahmadiyyah, Christians, LGBTQ+, Indonesian Chinese, and people with disabilities, as well as one additional group of interest: Jews. The insights gained from our dashboard have assisted stakeholders in devising more effective strategies to counteract hate speech. Notably, our dashboard has persuaded the General Election Supervisory Body in Indonesia (BAWASLU) to collaborate with our institution and the Alliance of Independent Journalists (AJI) to monitor social media hate speech in vulnerable areas in the country known for hate speech dissemination or hate-related violence in the upcoming Indonesian regional elections. This dashboard is available online at https://aji.or.id/hate-speech-monitoring.