A Late Fusion Approach Using CSNNs for Multi-Modal Toxicity Detection in Online Media
Bibliographic record
Abstract
With the growing prevalence of toxic behavior across online platforms, detecting harmful content spanning text, audio, and visual modalities has become an urgent challenge. This paper presents a comprehensive late fusion framework based on Convolutional Spiking Neural Networks (CSNNs), designed to capture complex temporal and cross-modal relationships for multi-modal toxicity detection. Our system integrates domain-adapted language models, mel spectrogram-derived audio features, and spatiotemporal visual cues, creating a synergistic architecture capable of handling inter-and intra-user variability. Extensive experiments conducted on a curated YouTube dataset comprising 931 annotated videos demonstrate that the proposed framework achieves superior performance compared to single-modality and early-fusion baselines, particularly in scenarios involving ambiguous or context-dependent toxic behavior. Beyond classification accuracy, the system’s adaptive design offers scalability for real-time deployment in content moderation pipelines. This work highlights the potential of combining bio-inspired spiking dynamics with late fusion strategies to address the evolving landscape of online toxicity and paves the way for future research on explainable, multi-lingual, and low-latency toxicity detection systems.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".