AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

p-MoD: Building Mixture-of-Depths MLLMs via Progressive Ratio Decay

Jun Zhang, Desen Meng, Zhengming Zhang, Zhenpeng Huang, Tao Wu, Limin Wang

ICCV 2025 · Vision-language and multimodal › Multimodal LLMs · approach: Training-free token reduction for MLLMs · 13 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 TanhNorm to vanilla MoD (ablation)p-MoD · AVG of 6 benchmarks (ablation)AVG: 60.4 -> 61.4Improved
add STRing symmetric token reweighting (ablation)p-MoD · AVG of 6 benchmarks (ablation)AVG: 61.4 -> 63.4Improved
add progressive ratio decay (PRD) (ablation)p-MoD · AVG of 6 benchmarks (ablation)AVG: 63.4 -> 65.3Improved
use TanhNorm instead of softmax weight normalization (ablation)p-MoD · AVG of 6 benchmarks (ablation)AVG: 62.9 -> 65.3Improved
use TanhNorm instead of shifted softmax (ablation)p-MoD · AVG of 6 benchmarks (ablation)AVG: 62.4 -> 65.3Improved
use gating factor alpha 0.2 instead of 1 in TanhNorm (ablation)p-MoD · AVG of 6 benchmarks (ablation)stability: overflow -> 65.3Improved
add weight normalization (TanhNorm) over none (ablation)p-MoD · AVG of 6 benchmarks (ablation)stability: overflow -> 65.3Improved
use decaying ratio schedules (cosine, linear) over constant/interleaved (ablation)p-MoD · AVG of 6 benchmarks (ablation)AVGImproved
use shifted cosine decay instead of stepped decay (ablation)p-MoD · AVG of 6 benchmarks (ablation)AVGImproved
raise shift factor beta from 0.3 to 0.5 (ablation)p-MoD · AVG of 6 benchmarks (ablation)AVG: 63.4 -> 65.3Improved
raise shift factor beta from 0.3 to 0.5 (ablation)p-MoD · AVG of 6 benchmarks (ablation)TFLOPs: 17.78 -> 21.94Hurt
apply p-MoD to LLaVA-NeXT-7Bp-MoD · AVG over 15 benchmarksTFLOPs: 39.46 -> 21.94Improved

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