analiza aiqbench
This research demonstrates how Large Language Models can significantly enhance hate speech detection by generating auxiliary context, improving F1 scores on both textual and multimodal datasets. The study evaluates specific fusion strategies, identifying embedding concatenation as the most effective method for integrating this generated background into SBERT-based classifiers. By addressing implicit bias in social media content, this work advances the practical application of LLMs in safety-critical NLP tasks.
Leveraging LLMs for Context-Aware Implicit Textual and Multimodal Hate Speech Detection
arXiv:2510.15685v2 Announce Type: replace Abstract: This paper investigates the use of an LLM to generate auxiliary background context for social media posts, and explores four methods to incorporate this context into the input of an SBERT-based Hate Speech Detection (HSD) classifier. These are: text concatenation, embedding concatenation, a hierarchical transformer-based fusion, and LLM-driven text enhancement. We evaluate the impact of our context generation and incorporation strategies in a textual setting on the Latent Hatred dataset of implicitly hateful tweets and a multimodal setting on the MAMI dataset of misogynous internet memes. Results are evaluated against a zero-context baseline, two previous approaches based on entity linking, and a zero-shot LLM classifier. Findings indicate that incorporating generated context improves HSD performance by up to 3 and 6 F1 points on textual and multimodal settings respectively, from a zero-context baseline to the highest-performing system, based on embedding concatenation.
Abstract:This paper investigates the use of an LLM to generate auxiliary background context for social media posts, and explores four methods to incorporate this context into the input of an SBERT-based Hate Speech Detection (HSD) classifier. These are: text concatenation, embedding concatenation, a hierarchical transformer-based fusion, and LLM-driven text enhancement. We evaluate the impact of our context generation and incorporation strategies in a textual setting on the Latent Hatred dataset of implicitly hateful tweets and a multimodal setting on the MAMI dataset of misogynous internet memes. Results are evaluated against a zero-context baseline, two previous approaches based on entity linking, and a zero-shot LLM classifier. Findings indicate that incorporating generated context improves HSD performance by up to 3 and 6 F1 points on textual and multimodal settings respectively, from a zero-context baseline to the highest-performing system, based on embedding concatenation.
Computation and Language (cs.CL)
arXiv:2510.15685 [cs.CL]
(or arXiv:2510.15685v2 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2510.15685 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 233 cuvinte
Subiecte conexe
Beyond a Joke: Multi-Angle Reasoning for Detecting and Explaining Harmful Humor in Memes
arXiv · 20 iul. 2026
The Continual Constitutional Convention: How language models revise the documents that govern them
AI Alignment Forum · 13 oct. 2026
How to Build In-Vehicle AI Agents with NVIDIA: From Cloud to Car
NVIDIA · 5 mai 2026
Multi-Granularity Context-Enhanced RAG over Multimodal Knowledge Graphs
arXiv · 27 aug. 2026