Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Learning Knowledge Graph-based World Models of Textual Environments
Prithviraj Ammanabrolu, Mark Riedl
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 graph difference prediction (ablation) | Worldformer · JerichoWorld KG prediction, overall | Graph F1: 31.41 -> 41.06 | Improved |
| add graph difference prediction (ablation) | Worldformer · JerichoWorld KG prediction, overall | Token F1: 41.02 -> 52.45 | Improved |
| add multi-task training (ablation) | Worldformer · JerichoWorld KG prediction, overall | Graph F1: 34.65 -> 41.06 | Improved |
| add multi-task training (ablation) | Worldformer · JerichoWorld KG prediction, overall | Token EM: 42.74 -> 51.32 | Improved |
| add SOS loss (ablation) | Worldformer · JerichoWorld KG prediction, overall | Graph F1: 36.17 -> 41.06 | Improved |
| add SOS loss (ablation) | Worldformer · JerichoWorld KG prediction, overall | Token EM: 48.82 -> 51.32 | Improved |
| add all three components to Seq2Seq baseline (ablation) | Worldformer · JerichoWorld KG prediction, overall | Graph EM: 14.29 -> 39.15 | Improved |
| add multi-task training (ablation) | Worldformer · JerichoWorld valid action prediction, overall | EM: 20.78 -> 23.22 | Improved |
| add multi-task training (ablation) | Worldformer · JerichoWorld valid action prediction, overall | F1: 22.42 -> 25.54 | Improved |
| add SOS loss (ablation) | Worldformer · JerichoWorld valid action prediction, overall | EM: 20.12 -> 23.22 | Improved |
| add SOS loss (ablation) | Worldformer · JerichoWorld valid action prediction, overall | F1: 21.28 -> 25.54 | Improved |
| add all components to Seq2Seq baseline (ablation) | Worldformer · JerichoWorld valid action prediction, overall | EM: 18.10 -> 23.22 | Improved |
Closest papers by meaning
- Learning to Model the World With Language (ICML 2024 · similarity 0.88)
- Monte Carlo Planning with Large Language Model for Text-Based Game Agents (ICLR 2025 · similarity 0.88)
- Reinforcement World Model Learning for LLM-based Agents (NeurIPS 2026 · similarity 0.88)
- AgentOdyssey: Open-Ended Long-Horizon Text Game Generation for Test-Time Continual Learning Agents (NeurIPS 2026 · similarity 0.88)
- TALES: Text-Adventure Learning Environment Suite (NeurIPS 2026 · similarity 0.87)
- Do LLMs Build World Representations? Probing Through the Lens of State Abstraction (NeurIPS 2024 · similarity 0.87)
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.
