The paper investigates whether large language models (LLMs) can generate synthetic cyberbullying conversations that replicate the social dynamics of real interactions. Using a comprehensive framework, the authors compare authentic dialogues with synthetic ones from GPT, Grok, and LLaMA across structural, linguistic, affective, and temporal dimensions, and conduct human evaluations of realism. Results show that while LLMs preserve high‑level interaction patterns, they systematically distort finer‑grained social phenomena, with model‑specific biases such as GPT’s suppression of harmful content and Grok’s amplification of aggression.
arXiv:2608. 11200v1 Announce Type: cross Abstract: Synthetic dialogue generation offers a way to study conversational dynamics in sensitive domains where real data are difficult to access, release, or annotate.
Synthetic dialogue generation offers a way to study conversational dynamics in sensitive domains where real data are difficult to access, release, or annotate. The underlying abuse may occur online or offline: threats and coercion can appear directly in messages, while behaviours such as surveillance, isolation, stalking, and physical violence may be planned, disclosed, or referred to conversationally.
The paper introduces a digital‑twin framework that simulates opinion dynamics in real Twitter networks by assigning agents attributes such as persona, emotions, centrality, stubbornness, and influence, and using Mistral‑7B to update opinions based on memory and social exposure. Validation on COVID‑19 and U.S. election 2020 datasets shows the framework reproduces opinion trajectories, reducing prediction error by over 50% compared to classical baselines, and improves structural alignment and polarization dynamics. Ablation studies reveal that agent attributes, memory, and social exposure all contribute to predictive fidelity, with agent attributes being the most critical.
arXiv:2609.13579v1 Announce Type: new Abstract: Safety research often focuses on model-generated harms, but users may also direct hostility, coercion, and adversarial pressure at models. Understandin...
The paper introduces CoCoEval, a framework for evaluating large language model (LLM)–simulated conversations by detecting 10 types of inconsistent and uncollaborative behaviors at the turn level. Using CoCoEval, the authors compare human conversations with those generated by GPT‑4.1, GPT‑5.1, and Claude Opus 4, finding that LLMs produce far fewer such behaviors under vanilla prompting and that prompt engineering or fine‑tuning often over‑produces specific behaviors. The study highlights gaps between human and LLM‑simulated interactions that conventional Likert‑scale evaluations miss, raising concerns about using LLMs as proxies for human social interaction.
The paper introduces BenCoDiC, a large Bengali cyberbullying corpus focused on political news discourse. It contains 50,250 textual instances from 100 YouTube videos, preserving caption–comment–reply relationships and organized via a nine‑dimensional star schema. A two‑tier human annotation process achieved high reliability (Fleiss' κ = 0.86, Krippendorff's α = 0.86), and transformer experiments show that adding conversational context modestly improves classification over isolated text.
arXiv:2510. 14207v3 Announce Type: replace Abstract: Large Language Model (LLM) agents are powering a growing share of interactive web applications, yet remain vulnerable to misuse and harm.
The paper presents a red‑teaming framework to evaluate how open‑source large language models (LLMs) can be used for political influence, focusing on models that can be deployed locally without weight modifications. It measures LLM Overton Windows (OWs) to assess expression fidelity of sampled political opinions and examines how natural‑language jailbreaks alter these scores. The study evaluates over 30 LLMs from 10 families across five countries, finding that left‑leaning content is expressed with higher fidelity, larger models tend to have lower mean OW scores, and jailbreak effectiveness varies significantly between model families.
The paper introduces FAID, a fine‑grained adaptive framework for detecting implicit hate speech. It first classifies samples into Shallow, Targeted, or Context‑Dependent categories and then applies tailored strategies—prompt‑tuning for shallow cases, knowledge augmentation for targeted ones, and an agentic prompt‑generation system for context‑dependent posts. Experiments on four benchmark datasets show that FAID outperforms state‑of‑the‑art baselines by allocating computational effort only where needed.
arXiv:2607. 20447v1 Announce Type: cross Abstract: This paper describes our system for the EEUCA 2026 Shared Task on toxicity classification in gaming chat.
arXiv:2608. 05050v1 Announce Type: cross Abstract: Against the backdrop of violence in police interactions with the U.
Against the backdrop of violence in police interactions with the U. S.