Research map / Architectures and efficiency / RNNs and normalization
Who Blames Whom in a Crisis? Detecting Blame Ties from News Articles Using Neural Networks
Shuailong Liang, Olivia Nicol, Yue Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| Add entity-prior features to the context model (ablation) | Combined model · ALL test | F1: 66.35 -> 69.92 | Improved |
| Add entity-prior features to the context model (ablation) | Combined model · KNOWN test | F1: 63.11 -> 68.67 | Hurt |
| Use context features rather than entity priors alone (ablation) | Blame tie extraction · ALL test | F1: 60.06 -> 66.35 | Improved |
| Use BiLSTM instead of unidirectional LSTM (ablation) | Context model · ALL dev | F1: 51.49 -> 62.26 | Improved |
| Use max pooling instead of random pooling (ablation) | BiLSTM context model · ALL dev | F1: 56.43 -> 62.26 | Improved |
| Use max pooling instead of mean pooling (ablation) | BiLSTM context model · ALL dev | F1: 61.95 -> 62.26 | Improved |
| Use max pooling instead of attention pooling (ablation) | BiLSTM context model · ALL dev | F1: 61.92 -> 62.26 | Improved |
| Use fixed pretrained GloVe vectors rather than random initialization (ablation) | BiLSTM + max pooling · ALL test | F1: 56.11 -> 62.10 | Improved |
| Fine-tune pretrained GloVe vectors rather than use random initialization (ablation) | BiLSTM + max pooling · ALL test | F1: 56.11 -> 57.75 | Improved |
| Use pretrained ELMo vectors rather than random initialization (ablation) | BiLSTM + max pooling · ALL test | F1: 56.11 -> 66.35 | Improved |
| Use entity-prior model on all pairs, including unseen entities | Entity model · ALL test | F1: 70.97 -> 60.06 | Hurt |
| Apply the context model to articles beyond the training period | Context model · Blame ties in new articles | F1: 72.00% | Improved |
Closest papers by meaning
- Local Identifying Causal Relations in the Presence of Latent Variables (ICML 2025 · similarity 0.79)
- Causal Discovery from Event Sequences by Local Cause-Effect Attribution (NeurIPS 2024 · similarity 0.79)
- Mastering Context-to-Label Representation Transformation for Event Causality Identification with Diffusion Models (AAAI 2024 · similarity 0.79)
- Causality can systematically address the monsters under the bench(marks) (NeurIPS 2026 · similarity 0.79)
- Causal Order: The Key to Leveraging Imperfect Experts in Causal Inference (ICLR 2025 · similarity 0.78)
- MOCHA: Discovering Multi-Order Dynamic Causal Structure in Temporal Point Processes (NeurIPS 2026 · similarity 0.78)
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.
