AI Paper Atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

Tighter Truncated Rectangular Prism Approximation for RNN Robustness Verification

Xingqi Lin, Liangyu Chen, Min Wu, Min Zhang, Zhenbing Zeng

AAAI 2026 · Learning theory and algorithms › Combinatorial and search optimization · approach: Semidefinite relaxations for verification · 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
replace distance-based LP planes with volume-area hybrid objectiveProver · MNIST (eps=0.012, f=4, h=32, layers=1)verified accuracy (%): 45 -> 83Improved
replace distance-based LP planes with volume-area hybrid objectiveProver · MNIST (eps=0.012, f=4, h=32, layers=3)verified accuracy (%): 25 -> 59Improved
replace distance-based LP planes with volume-area hybrid objectiveProver · MNIST (eps=0.012, f=4, h=128, layers=1)verified accuracy (%): 86 -> 93Improved
volume-area objective cost in single-plane verificationProver · MNIST (eps=0.012, f=4, h=32, layers=1)running time (s): 1.47 -> 1.79Hurt
volume-area objective on speech datasetsProver · GSC (eps=-100 dB)verified accuracy (%): 44 -> 46Improved
volume-area objective on sentiment analysisProver · RT sentiment (eps=0.05)verified accuracy (%): 77 -> 84Improved
volume-area objective on sentiment analysis, high perturbationProver · RT sentiment (eps=0.09)verified accuracy (%): 27 -> 28No clear effect
use 4-rectangle division instead of 2-triangle (up) in multi-plane (ablation)DeepPrism · MNIST multi-plane (eps=0.020)verified accuracy (%): 18 -> 54Improved
use finer 16-rectangle division instead of 4-rectangle (ablation)DeepPrism · MNIST multi-plane (eps=0.020)verified accuracy (%): 54 -> 64Improved
multi-plane refinement vs single-plane (ablation)DeepPrism · MNIST (f=4, h=32, layers=3)verified accuracy (%)Improved
use different LP approximation objective inside multi-plane refinement (ablation)multi-plane approximation · MNIST (f=4, h=32, layers=3)verified accuracy (%)No clear effect

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