AI Paper Atlas

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

From Pixels to UI Actions: Learning to Follow Instructions via Graphical User Interfaces

Peter Shaw, Mandar Joshi, James Cohan, Jonathan Berant, Panupong Pasupat, Hexiang Hu, Urvashi Khandelwal, Kenton Lee, Kristina Toutanova

NeurIPS 2023 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Imitation and skill learning · 8 citations (OpenAlex)

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 changedWhereMeasuredEffect
add Pix2Struct pre-training (BC) (ablation)Pix2Act · MiniWob++mean score: 17.1 -> 66.5Improved
add Pix2Struct pre-training (BC) (ablation)Pix2Act · WebShoptask score: 1.1 -> 46.7Improved
pre-training on held-out tasks (ablation)Pix2Act · MiniWob++ 9 held-out tasksmean score: 7.6 -> 28.3Improved
pre-training when tasks are in training set (ablation)Pix2Act · MiniWob++ 9 tasksmean score: 11.0 -> 65.5Improved
intermediate finetuning on MiniWob++ before WebShop (ablation)Pix2Act · WebShoptask score: drop of 4.0 without itImproved
tree search policy improvement iteration 0 to 1 (greedy)Pix2Act · MiniWob++mean score: 66.5 -> 93.1Improved
second iteration of tree search training (greedy)Pix2Act · MiniWob++mean score: 93.1 -> 96.2Improved
use tree search at inference instead of greedy, iteration 0Pix2Act · MiniWob++mean score: 66.5 -> 91.7Improved
use tree search at inference instead of greedy, iteration 1Pix2Act · MiniWob++mean score: 93.1 -> 98.4Improved
rollout only tree search vs greedy (ablation)Pix2Act · MiniWob++ 12 challenging tasksmean score: 28.8 -> 74.2Improved
add value network to rollouts (lambda=0.1) (ablation)tree search · MiniWob++ 12 challenging tasksmean score: 74.2 -> 78.3Improved
value network only (lambda=1) instead of mixed (ablation)tree search · MiniWob++ 12 challenging tasksmean score: 78.3 -> 57.4Hurt

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