Research map / Architectures and efficiency / Attention and transformers
SAM-Net: Integrating Event-Level and Chain-Level Attentions to Predict What Happens Next
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 |
|---|---|---|---|
| add self attention (event segments) to SAM-Net (ablation) | SAM-Net · MCNC NYT test | accuracy (%): 49.54 -> 54.48 | Helped |
| add chain-level attention (ablation) | SAM-Net · MCNC NYT test | accuracy (%): 51.38 -> 54.48 | Helped |
| add event-level attention (ablation) | SAM-Net · MCNC NYT test | accuracy (%): 53.25 -> 54.48 | Helped |
| add DenseNet feature extraction instead of averaging (ablation) | SAM-Net · MCNC NYT test | accuracy (%): 52.96 -> 54.48 | Helped |
| use richer event representation (v, subject, object, prep) instead of <verb, dependency> (ablation) | SAM-Net · MCNC NYT test | accuracy (%): 52.35 -> 54.48 | Helped |
| Fusion similarity beta instead of cosine similarity (ablation) | SAM-Net · MCNC NYT test | accuracy (%): 53.23 -> 54.48 | Helped |
| Fusion similarity beta instead of dot similarity (ablation) | SAM-Net · MCNC NYT test | accuracy (%): 53.45 -> 54.48 | Helped |
| Fusion similarity beta instead of fusion alpha (ablation) | SAM-Net · MCNC NYT test | accuracy (%): 54.22 -> 54.48 | Helped |
| SAM-Net versus best baseline SGNN | SAM-Net · MCNC NYT test | accuracy (%): 52.32 -> 54.48 | Helped |
| neural event embeddings versus count-based PMI | Word2vec · MCNC NYT test | accuracy (%): 31.25 -> 42.10 | Helped |
| LSTM order modeling alone versus pairwise Event-Comp | LSTM · MCNC NYT test | accuracy (%): 49.66 -> 46.75 | Hurt |
Closest papers by meaning
- A Generative Approach for Script Event Prediction via Contrastive Fine-Tuning (AAAI 2023 · similarity 0.90)
- Rich Event Modeling for Script Event Prediction (AAAI 2023 · similarity 0.88)
- FEEL: Featured Event Embedding Learning (AAAI 2018 · similarity 0.87)
- What Happens Next? Event Prediction Using a Compositional Neural Network Model (AAAI 2016 · similarity 0.87)
- Event Representations With Tensor-Based Compositions (AAAI 2018 · similarity 0.86)
- Don't Take Things Out of Context: Attention Intervention for Enhancing Chain-of-Thought Reasoning in Large Language Models (ICLR 2025 · similarity 0.85)
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.