A new multilingual dataset and an SFT+DPO LLM pipeline for generating comment aspect terms yields small clustering improvements, but the benchmark construction and missing artifacts limit the strength of the evidence.
Deep Learning for User Comment Moderation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
Experimenting with a new dataset of 1.6M user comments from a Greek news portal and existing datasets of English Wikipedia comments, we show that an RNN outperforms the previous state of the art in moderation. A deep, classification-specific attention mechanism improves further the overall performance of the RNN. We also compare against a CNN and a word-list baseline, considering both fully automatic and semi-automatic moderation.
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Improving Multilingual Social Media Insights: Aspect-based Comment Analysis
A new multilingual dataset and an SFT+DPO LLM pipeline for generating comment aspect terms yields small clustering improvements, but the benchmark construction and missing artifacts limit the strength of the evidence.