Research map / Large language models / Language, safety and interpretability
Violence Rating Prediction from Movie Scripts
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use full feature set instead of word2vec-only semantic features (ablation) | GRU-16 · Violence rating | macro F1 (%): 51.5 -> 60.4 | Helped |
| add sentiment features (ablation) | GRU-16 · Violence rating | macro F1 (%): 58.8 -> 60.4 | Helped |
| add semantic (word2vec) features (ablation) | GRU-16 · Violence rating | macro F1 (%): 58.9 -> 60.4 | Helped |
| add movie genre feature (ablation) | GRU-16 · Violence rating | macro F1 (%): 59.4 -> 60.4 | Helped |
| add N-gram features (ablation) | GRU-16 · Violence rating | macro F1 (%): 59.5 -> 60.4 | Helped |
| add abusive language features (ablation) | GRU-16 · Violence rating | macro F1 (%): 59.6 -> 60.4 | Helped |
| add linguistic features (ablation) | GRU-16 · Violence rating | macro F1 (%): 60.1 -> 60.4 | No clear effect |
| use TF-IDF instead of TF for n-grams (ablation) | GRU-16 · Violence rating | macro F1 (%): 60.4 -> 56.9 | Hurt |
| use paragraph2vec instead of averaged word2vec (ablation) | GRU-16 · Violence rating | macro F1 (%): 60.4 -> 59.1 | Hurt |
| increase n-gram vocabulary size from 500 to 5000 (ablation) | GRU-16 · Violence rating | macro F1 (%): 55.9 -> 60.4 | Helped |
| train on last act instead of first segment (ablation) | GRU-16 · Violence rating | macro F1 (%) | Helped |
| GRU-16 vs movie-level Linear SVC | violence classifier · Violence rating | macro F1 (%): 59.1 -> 60.4 | No clear effect |
Closest papers by meaning
- A Multi-Task Benchmark for Abusive Language Detection in Low-Resource Settings (NeurIPS 2025 · similarity 0.82)
- Evaluating Multimodal Narrative Understanding of Popular Hollywood Films (NeurIPS 2026 · similarity 0.80)
- MovieBench: A Hierarchical Movie Level Dataset for Long Video Generation (CVPR 2025 · similarity 0.80)
- ShotBench: Expert-Level Cinematic Understanding in Vision-Language Models (NeurIPS 2025 · similarity 0.80)
- Infer Induced Sentiment of Comment Response to Video: A New Task, Dataset and Baseline (NeurIPS 2024 · similarity 0.79)
- CineTechBench: A Benchmark for Cinematographic Technique Understanding and Generation (NeurIPS 2025 · similarity 0.79)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.