Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Language as a Cognitive Tool to Imagine Goals in Curiosity Driven Exploration
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 |
|---|---|---|---|
| enable goal imagination (compose known goals) (ablation) | IMAGINE with modular-attention policy · Playground testing goals | SR_test: 0.51 -> 0.76 | Helped |
| enable goal imagination (ablation) | IMAGINE with flat-architecture policy · Playground testing goals | SR_test: 0.17 -> 0.15 | No clear effect |
| use modular-attention policy instead of flat architecture (ablation) | IMAGINE · Playground testing goals | SR_test: 0.17 -> 0.51 (no imagination); 0.15 -> 0.76 (imagination) | Helped |
| reduce precision of imagined goals (ablation) | goal imagination · Playground testing goals | SR_test | Hurt |
| reduce coverage of imagined goals (ablation) | goal imagination · Playground testing goals | SR_test | No clear effect |
| replace construction-grammar imagination with random word sequences (ablation) | goal imagination · Playground testing goals | SR_test | No clear effect |
| social partner present only 10% of episodes (spread uniformly), with imagination (ablation) | IMAGINE · Playground testing goals | SR_test | Hurt |
| concentrate same 10% feedback in first 10% of episodes, with imagination (ablation) | IMAGINE · Playground testing goals | SR_test | Helped |
Closest papers by meaning
- SENSEI: Semantic Exploration Guided by Foundation Models to Learn Versatile World Models (ICML 2025 · similarity 0.87)
- Language Model Goal Selection Differs from Humans' in a Self-Directed Learning Task (NeurIPS 2026 · similarity 0.86)
- GoalLadder: Incremental Goal Discovery with Vision-Language Models (NeurIPS 2025 · similarity 0.86)
- A Single Goal is All You Need: Skills and Exploration Emerge from Contrastive RL without Rewards, Demonstrations, or Subgoals (ICLR 2025 · similarity 0.86)
- OMNI-EPIC: Open-endedness via Models of human Notions of Interestingness with Environments Programmed in Code (ICLR 2025 · similarity 0.85)
- Do Visual Imaginations Improve Vision-and-Language Navigation Agents? (CVPR 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.