Research map / Large language models / Retrieval and knowledge
Active Reasoning in an Open-World Environment
Manjie Xu, Guangyuan Jiang, Wei Liang, Chi Zhang, Yixin Zhu
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 |
|---|---|---|---|
| use the Ideal explorer instead of the TRPO explorer (ablation) | FrozenBiLM-DeBERTa · Active reasoning QA overall | accuracy (%): 65.3 -> 77.8 | Improved |
| use the Ideal explorer instead of the TRPO explorer (ablation) | Flamingo-Mini · Active reasoning QA overall | accuracy (%): 66.3 -> 69.0 | Improved |
| use the Ideal explorer instead of the TRPO explorer (ablation) | Vanilla-Trans · Active reasoning QA overall | accuracy (%): 28.8 -> 66.1 | Improved |
| use the Ideal explorer instead of the TRPO explorer (ablation) | FrozenBiLM-BERT · Active reasoning QA overall | accuracy (%): 61.0 -> 74.0 | Improved |
| use the Ideal explorer instead of the TRPO explorer (ablation) | FrozenBiLM-DeBERTa · Goal questions | accuracy (%): 43.1 -> 71.8 | Improved |
| replace TRPO explorer with empty visual inputs (ablation) | FrozenBiLM-BERT · Goal questions | accuracy (%): 44.4 -> 25.5 | Hurt |
| replace TRPO explorer with empty visual inputs (ablation) | Flamingo-Mini · Goal questions | accuracy (%): 43.3 -> 22.9 | Hurt |
| use AFD (deductive simulation) instead of the standard setting (ablation) | FrozenBiLM-BERT · Active reasoning QA overall | accuracy (%): 61.0 -> 66.0 | Improved |
| use AFD instead of the standard setting (ablation) | FrozenBiLM-DeBERTa · Active reasoning QA overall | accuracy (%): 65.3 -> 65.4 | No clear effect |
| use AFD instead of the standard setting (ablation) | Flamingo-Mini · Active reasoning QA overall | accuracy (%): 66.3 -> 66.1 | No clear effect |
| use AFD instead of the standard setting (ablation) | FrozenBiLM-BERT · Survival questions | accuracy (%): 54.4 -> 55.5 | Improved |
| use AFD instead of the standard setting (ablation) | Vanilla-Trans · Active reasoning QA overall | accuracy (%): 28.8 -> 24.3 | Hurt |
Closest papers by meaning
- From Passive to Active Reasoning: Can Large Language Models Ask the Right Questions under Incomplete Information? (ICML 2025 · similarity 0.89)
- Mars: Situated Inductive Reasoning in an Open-World Environment (NeurIPS 2024 · similarity 0.88)
- Eliciting Reasoning in Language Models with Cognitive Tools (NeurIPS 2025 · similarity 0.86)
- Active Exploring like a Pigeon: Reinforcing Spatial Reasoning via Agentic Vision-Language Models (ICML 2026 · similarity 0.86)
- PuzzleWorld: A Benchmark for Multimodal, Open-Ended Reasoning in Puzzlehunts (ICLR 2026 · similarity 0.85)
- PhysVLM-AVR: Active Visual Reasoning for Multimodal Large Language Models in Physical Environments (NeurIPS 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.
