Research map / Large language models / AI and society
Atari-HEAD: Atari Human Eye-Tracking and Demonstration Dataset
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 predicted gaze saliency to the action policy | AtariHead-IL · Alien | game score: +112.33% | Helped |
| Add predicted gaze saliency to the action policy | AtariHead-IL · Asterix | game score: +43.96% | Helped |
| Add predicted gaze saliency to the action policy | AtariHead-IL · Breakout | game score: +1138.46% | Helped |
| Add predicted gaze saliency to the action policy | AtariHead-IL · Enduro | game score: -5.53% | Hurt |
| Add predicted gaze saliency to the action policy | AtariHead-IL · Freeway | game score: +0.33% | No clear effect |
| Add predicted gaze saliency to the action policy | AtariHead-IL · Hero | game score: +0.22% | No clear effect |
| Add predicted gaze saliency to the action policy | AtariHead-IL · Montezuma's Revenge | game score: +104.05% | Helped |
| Add predicted gaze saliency to the action policy | AtariHead-IL · Seaquest | game score: +309.05% | Helped |
| Add predicted gaze saliency to the action policy | AtariHead-IL · Space Invaders | game score: +0.45% | No clear effect |
| Add predicted gaze saliency to the action policy | AtariHead-IL · 20 Atari games | game score: 115.26% average improvement | Mixed |
| Use a gaze-guided policy rather than standard behavior cloning | AtariHead-IL · Alien | behavior matching accuracy: 0.504 -> 0.690 | Helped |
| Use a gaze-guided policy rather than standard behavior cloning | AtariHead-IL · Asterix | behavior matching accuracy: 0.410 -> 0.532 | Helped |
Closest papers by meaning
- EgoExoLearn: A Dataset for Bridging Asynchronous Ego- and Exo-centric View of Procedural Activities in Real World (CVPR 2024 · similarity 0.85)
- GameVerse: Can Vision-Language Models Learn from Video-based Reflection? (ICML 2026 · similarity 0.85)
- Investigating Pre-Training Objectives for Generalization in Vision-Based Reinforcement Learning (ICML 2024 · similarity 0.84)
- EyEar: Learning Audio Synchronized Human Gaze Trajectory Based on Physics-Informed Dynamics (AAAI 2025 · similarity 0.84)
- Attention Trajectories as a Diagnostic Axis for Deep Reinforcement Learning (NeurIPS 2026 · similarity 0.84)
- Diffusion for World Modeling: Visual Details Matter in Atari (NeurIPS 2024 · 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.