Research map / Vision-language and multimodal / Human and embodied modeling
Predicting Event Memorability from Contextual Visual Semantics
Qianli Xu, Fen Fang, Ana Garcia del Molino, Vigneshwaran Subbaraju, Joo Hwee Lim
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 extrinsic features via MLP pipeline to AMNet (ablation) | AMNet · R3 dataset | F1: 0.150 -> 0.368 | Improved |
| add extrinsic features via MLP pipeline to ICNet (ablation) | ICNet · R3 dataset | F1: 0.140 -> 0.340 | Improved |
| add intrinsic image features (AMNet) to MLP (ablation) | MLP · R3 dataset | F1: 0.333 -> 0.368 | Improved |
| add intrinsic image features (ICNet) to MLP (ablation) | MLP · R3 dataset | F1: 0.333 -> 0.340 | No clear effect |
| use extrinsic-only MLP instead of image CNN | MLP vs AMNet · R3 dataset | F1: 0.150 -> 0.333 | Improved |
| use AMNet instead of ICNet in pipeline 2 | CEMNet · R3 dataset | F1: 0.340 -> 0.368 | Improved |
| use CEMNet w/t AMNet versus image-only AMNet | AMNet · R3 dataset | mean absolute error: 3.03 -> 0.85 | Improved |
| use pre-trained MemNet image memorability to predict event memory | MemNet · R3 dataset | Pearson r: r = 0.02 | No clear effect |
| use pre-trained DeepNSM image memorability | DeepNSM · R3 dataset | Pearson r: r = 0.01 | No clear effect |
| use pre-trained AMNet image memorability | AMNet · R3 dataset | Pearson r: r = 0.19 | Improved |
| remove activity, place, face, boundary, test or encode distinctiveness features (ablation) | CEMNet · R3 dataset | F1 | Improved |
| remove human, treatment or interval features (ablation) | CEMNet · R3 dataset | F1 | Mixed |
Closest papers by meaning
- Episodic Memories Generation and Evaluation Benchmark for Large Language Models (ICLR 2025 · similarity 0.88)
- Human-inspired Episodic Memory for Infinite Context LLMs (ICLR 2025 · similarity 0.87)
- Quantifying Cross-Modality Memorization in Vision-Language Models (NeurIPS 2025 · similarity 0.84)
- VisMem: Latent Vision Memory Unlocks Potential of Vision-Language Models (CVPR 2026 · similarity 0.84)
- REMem: Reasoning with Episodic Memory in Language Agent (ICLR 2026 · similarity 0.84)
- Titans: Learning to Memorize at Test Time (NeurIPS 2025 · similarity 0.83)
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.
