AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

AD-BTS: Adaptive Dual-Branch Token Sparsification via Spatial Information Density

Xinpei Gao, Xin Luo, Ming Liu, Chunjiang Wang, S Kevin Zhou

ICML 2026 · Architectures and efficiency › Pruning and efficient tuning · approach: Layer-wise pruning of LLMs

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
route inputs adaptively between RSB and SFB (delta=0.4) instead of RSB only (ablation)AD-BTS · ChartQA (30% tokens)Relaxed accuracy: 73.12 -> 83.68Improved
route inputs adaptively instead of SFB only (ablation)AD-BTS · MME (30% tokens)score: 2233.70 -> 2351.76Improved
dual-branch routing at delta=0.4 versus single-branch variants (ablation)AD-BTS · Nine image benchmarks, 30% tokensnormalized avg score (%): 97.11% / 98.01% -> 98.72%Improved
add LoRA adapters (r=256) in SFB versus r=0 (ablation)AD-BTS · ChartQA (30% tokens)Relaxed accuracy: 80.52 -> 83.68Improved
increase LoRA rank from 256 to 512 (ablation)AD-BTS · Nine image benchmarks, 30% tokensnormalized avg score (%): 98.72% -> 96.82%Hurt
lower LoRA rank to 64 (ablation)AD-BTS · Nine image benchmarks, 30% tokensnormalized avg score (%): 98.72% -> 97.50%Hurt
add context fusion on top of SFB with LoRA (ablation)AD-BTS · ChartQA (30% tokens)Relaxed accuracy: 79.96 -> 83.68Improved
use dual-branch routing instead of always-RSB with persistent LoRA at matched capacity (ablation)AD-BTS · ChartQA (30% tokens)Relaxed accuracy: 78.64 -> 83.68Improved
AD-BTS versus VisionSelector at extreme compressionQwen2.5-VL-7B (10% tokens) · ChartQARelaxed accuracy: 61.6 -> 68.24Improved
AD-BTS versus uncompressed model on videoQwen2.5-VL-7B · Video prefill timeprefill time (ms): 1413.34 -> 746.25Improved
RSB-only routing on redundancy-dominated MME (ablation)AD-BTS · MME (30% tokens)score: 2351.76 -> 2355.26Improved

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