atlas

Research map / Multi-agent systems and games / Human-AI and goal-conditioned

Visual Room Rearrangement

CVPR 2021 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Embodied agent learning · 95 citations (OpenAlex)

Check 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 changedWhereMeasuredEffect
Replace the Simple CNN backbone with pretrained ResNet18 (ablation)1-Phase IL · RoomR test100 · Success: 1.8 -> 2.8Helped
Replace the Simple CNN backbone with pretrained ResNet18 (ablation)1-Phase IL · RoomR test100 · %FixedStrict: 4.8 -> 6.3Helped
Add ANM semantic mapping (ablation)1-Phase (RN18, IL) · RoomR test100 · %FixedStrict: 6.3 -> 8.9Helped
Add ANM semantic mapping (ablation)2-Phase (RN18, PPO+IL) · RoomR test100 · %FixedStrict: 0.7 -> 1.4Helped
Use the sequential 2-Phase task rather than the aligned-image 1-Phase taskResNet18 models · RoomR test100 · Success: 2.8 -> 0.2Hurt
Train with PPO rather than imitation learning1-Phase ResNet18 · RoomR test100 · Success: 2.8 -> 1.1Hurt
Add a pretrained CNN backbone instead of the Simple CNN (ablation)1-Phase PPO · RoomR test100 · Success: 0.7 -> 1.1Mixed

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy