AI Paper Atlas

Research map / Point clouds / Point cloud classification

Point-Focused Attention Meets Context-Scan State Space: Robust Biological Visual Perception for Point Cloud Representation

Kanglin Qu, Pan Gao, Qun Dai, Yuanhao Sun

ICLR 2026 · Point clouds › Point cloud classification · approach: Local feature aggregation networks · 0 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 local neighbor branch to point-focused attention (ablation)PointLearner · ModelNet40OA: 92.11 -> 94.17Improved
add spatial downsampling branch (ablation)PointLearner · ModelNet40OA: 93.06 -> 94.17Improved
competitive single-softmax fusion instead of additive fusion (ablation)PointLearner · ModelNet40OA: 93.43 -> 94.17Improved
bidirectional S6 instead of unidirectional S6 (ablation)PointLearner · ModelNet40OA: 93.08 -> 94.17Improved
add context-scan state space on top of point-focused attention (ablation)PointLearner · ModelNet40OA: 92.93 -> 94.17Improved
add point-focused attention to context-scan state space (ablation)PointLearner · ModelNet40OA: 91.94 -> 94.17Improved
Hilbert serialization instead of none (ablation)PointLearner · ModelNet40OA: 91.34 -> 94.17Improved
Hilbert instead of Z-Order curve (ablation)PointLearner · ModelNet40OA: 93.06 -> 94.17Improved
combine two serialization curves instead of Hilbert alone (ablation)PointLearner · ModelNet40OA: 94.17 -> 93.78 (Hilbert & Trans-Hilbert), 93.52 (Hilbert & Z-Order)Hurt
learnable serialization instead of Hilbert curve (ablation)PointLearner · ModelNet40OA: 94.17 -> 92.78Hurt
use induced point pooling instead of FPS for spatial downsampling at high rate (/256) (ablation)PointLearner · ModelNet40OA: IPP 93.39 at /256; FPS approaches no-branch baselineImproved

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