AI Paper Atlas

Research map / Video / Video-language understanding

Reward Modeling for Computer-Using Agent from Video Execution

Linxin Song, Jieyu Zhang, Huanxin Sheng, Taiwei Shi, Rahul Gupta, Yang Liu, Ranjay Krishna, Jian Kang, Jieyu Zhao

ECCV 2026 · Video › Video-language understanding · approach: Event and anomaly reasoning in video

Read the paper

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
fine-tune on ExeVR-53k execution videosQwen3-VL 8B · ExeVR-Benchaccuracy: 67.7 -> 84.7Improved
use dense execution video instead of sparse screenshots (ablation)Qwen3-VL · ExeVR-Bench Ubuntu (Human)accuracy: AER 58.5 -> ExeVRM 4B 81.0Improved
reduce resolution to 360p to fit all key frames (ablation)SE-WSM / ZeroGUI · ExeVR-Bench Ubuntu (Human)accuracy: SE-WSM 48.5 vs AER 55.0 (8B)Hurt
raise input resolution from 360p to 720p with STP+TTP (ablation)ExeVRM 8B · ExeVR-Benchaccuracy: 81.5 -> 84.7Improved
raise input resolution from 360p to 720p with STP+TTP (ablation)ExeVRM 4B · ExeVR-Benchaccuracy: 79.3 -> 80.1; recall 77.8 -> 82.5; precision 80.6 -> 79.2Mixed
apply temporal token pruning alone vs spatial alone (ablation)ExeVRM 4B · ExeVR-Benchaccuracy: STP only 77.6 -> TTP only 80.3Improved
add STP on top of TTP (ablation)ExeVRM 4B · ExeVR-Benchrecall: accuracy 80.3 -> 80.1; recall 79.3 -> 82.5; precision 81.3 -> 79.2Mixed
combine STP and TTP for efficiency (ablation)ExeVRM 4B · 50 framespeak GPU memory: TTP-only ~68 GiB -> STP+TTP ~49.5 GiBImproved
scale ExeVRM from 4B to 8B (ablation)ExeVRM · ExeVR-Benchaccuracy: 80.1 -> 84.7Improved
train reward model on execution video for temporal attributionExeVRM 8B · ExeVR-BenchtIoU: Gemini 3 Pro 0.2494 -> 0.3332Improved

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