AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

OmniStream: Mastering Perception, Reconstruction and Action in Continuous Streams

Yibin Yan, Jilan Xu, Shangzhe Di, Haoning Wu, Weidi Xie

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Spatial reasoning in VLMs · 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 video data to SSL (ablation)OmniStream · SSv2accuracy: 63.0 -> 69.3Improved
add video data to SSL (ablation)OmniStream · DAVIS'17J&F: 67.7 -> 71.6Improved
add video data to SSL (ablation)OmniStream · CALVINavg. length: 3.42 -> 3.80Improved
add video data to SSL (ablation)OmniStream · VSI-Benchaccuracy: 57.9 -> 57.3Hurt
add streaming geometric reconstruction (ablation)OmniStream · NYUv2RMSE: 0.471 -> 0.379Improved
add streaming geometric reconstruction (ablation)OmniStream · ADE20KmIoU: 42.3 -> 49.6Improved
add streaming geometric reconstruction (ablation)OmniStream · VSI-Benchaccuracy: 52.5 -> 57.3Improved
add streaming geometric reconstruction (ablation)OmniStream · CALVINavg. length: 3.34 -> 3.80Improved
add captioning vision-language alignment (ablation)OmniStream · VideoMMEaccuracy: 45.0 -> 54.1Improved
add captioning vision-language alignment (ablation)OmniStream · VSI-Benchaccuracy: 44.9 -> 57.3Improved
add captioning vision-language alignment (ablation)OmniStream · CALVINavg. length: 2.38 -> 3.80Improved
use causal KV-cache instead of bidirectional recomputation (ablation)OmniStream · per-frame latency at T=64 (s)latency: 0.998 -> 0.067Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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