AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Adaptive DNA Sequence Modeling via Synergistic Plasticity Units

Binghao Liu, Wenzheng Zhao, Zhijie Zheng, Fei Gu

ICML 2026 · Architectures and efficiency › Attention and transformers · approach: State space and recurrent sequence models

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 LPL (ablation)SPU-DNA · Genomic Benchmark: Human vs. Wormaccuracy: 94.67 -> 97.19Improved
add DPL (ablation)SPU-DNA · Genomic Benchmark: Human vs. Wormaccuracy: 90.01 -> 97.19Improved
add SPL (ablation)SPU-DNA · Genomic Benchmark: Human vs. Wormaccuracy: 95.22 -> 97.19Improved
add DPL (ablation)SPU-DNA · Genomic Benchmark: Human Enhancers Ensemblaccuracy: 85.91 -> 91.08Improved
add LPL (ablation)SPU-DNA · Genomic Benchmark: Human Nontata Promotersaccuracy: 91.05 -> 94.78Improved
add SPL (ablation)SPU-DNA · Genomic Benchmark: Mouse Enhancersaccuracy: 81.02 -> 82.38Improved
SPU-DNA vs strongest baselines on Genomic BenchmarkSPU-DNA · Genomic Benchmark (8 datasets)accuracy: best on 7 of 8; Nontata Promoters 94.78 vs ConvNova 95.21Mixed
SPU-DNA vs baselines (<5M params)SPU-DNA · NT Benchmark (18 datasets)MCC/F1/accuracy: best on 16 of 18Mixed
scale SPU-DNA to 500M paramsSPU-DNA · NT BenchmarkMCC/F1/accuracy: leads on all datasets vs 1.9M model, e.g. Enhancer types 51.17 -> 62.98Improved
SPU-DNA (6.5M) vs ConvNova (27.4M)SPU-DNA · Chromatin Profile PredictionAUROC: DHS 93.71 -> 94.01Improved
replace SENet and Hyena gating with SPL (ablation)SPU-DNA · Information Bottleneck analysispredictive power and compressionImproved
SPU vs Transformer + FlashAttn at 100k lengthSPU · Efficiency, trainingVRAM and wall clock: wall clock 4.87s -> 0.93s; VRAM 7.8 GB -> 6.8 GBImproved

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