AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

RF-MatID: Dataset and Benchmark for Radio Frequency Material Identification

Xinyan Chen, Qinchun Li, Ruiqin Ma, Jiaqi Bai, Li Yi, Jianfei Yang

ICLR 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors · 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
shift test distances (cross-distance split) instead of random splitbaseline model · RF-MatIDaccuracy (%): 99.57 -> 69.47 (P1 S2-2)Hurt
shift test angles (cross-angle split) instead of random splitbaseline model · RF-MatIDaccuracy (%): 99.57 -> 85.23 (P1 S3-2)Hurt
use millimeter-wave band (P2) instead of centimeter-wave (P3)baseline model · RF-MatID cross-distanceaccuracy (%): 60.07 -> 62.31 (S2-2)Improved
use centimeter-wave band (P3) instead of millimeter-wave (P2)baseline model · RF-MatID cross-angleaccuracy (%): 75.17 -> 87.17 (S3-2)Improved
restrict to country-specific legal bands (P4, P5) instead of full spectrumbaseline model · RF-MatID S1accuracy (%): 99.57 -> 99.53 (P4), 99.41 (P5)No clear effect
use coarser superclass labels instead of 16 fine-grained classesbaseline model · RF-MatID P1 S2 mod1accuracy (%): 86.62 -> 91.15Improved
use sequence-oriented models (Transformer, BiLSTM, TimesNet)RF models · RF-MatIDaccuracy (%): BiLSTM 86.33 vs MLP 99.19 (P1 S1)Hurt
use frequency-domain input instead of time-domain inputbaseline model · RF-MatID P1accuracy (%)No clear effect
use classical background-subtraction method mSensemSense · RF-MatID P1 S1accuracy (%): 99.57 -> 10.31Hurt

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