AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Efficient Decoder Scaling Strategy for Constructive Neural Routing Solvers

Qing Luo, Fu Luo, Ke LI, Zhenkun Wang

NeurIPS 2026 · Architectures and efficiency › RNNs and normalization · approach: Efficient network compression and training

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Studies scaling the decoder size of construction-based neural routing solvers for vehicle routing problems, rather than only the encoder.

Problem

Construction-based neural routing solvers for vehicle routing typically use small decoders (1-3M parameters), leaving the effects of decoder scaling largely unexplored.

Approach

Systematic study of decoder scaling in encoder-decoder constructive routing solvers, shifting parameters toward the decoder and examining how performance scales with decoder size, with an efficiency-oriented scaling strategy.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
restructure to deep-narrow (D=96, W=32) at fixed ~1.3M parameters (ablation)Decoder-only (D=6,W=128) · TSP100optimality gap (%): 0.464% -> 0.184%Improved
restructure to deep-narrow (D=96, W=32) at fixed parameters (ablation)Decoder-only (D=6,W=128) · TSP1000optimality gap (%): 2.172% -> 0.991%Improved
increase depth 6 -> 42 instead of width, limited training data (0.1M) (ablation)Decoder-only · TSP100optimality gap (%): 1.715% -> 0.547%Improved
increase width 128 -> 512 with limited data (0.1M) (ablation)Decoder-only · TSP100optimality gap (%): 1.715% -> 0.949%Improved
scale depth instead of width in parameter scaling law (ablation)Decoder-only · TSP100scaling exponent: depth 0.98-1.05 vs width 0.24-0.40Improved
scale depth (6 -> 42) for data efficiency exponent (ablation)Decoder-only (W=128) · TSP100data scaling exponent: 0.47 -> 0.71Improved
scale depth 6 -> 42 for compute efficiency (ablation)Decoder-only (W=128) · TSP100 beam searchcompute exponent: 0.82 -> 1.71Improved
use medium-depth (24) model instead of wide model under 12-minute budget (ablation)Decoder-only · TSP100optimality gap (%): 0.0138 -> 0.0044Improved
use D=42 instead of D=24 with sufficient inference budget (ablation)Decoder-only (W=128) · TSP100optimality gap (%): 0.0002 -> 0.0001Improved
use deep model (42,128) with greedy decoding vs best baseline with beam16Decoder-only · TSP1000optimality gap (%): 1.35 -> 0.869Improved
fit single parameter-count power law to all 12 models (ablation)Decoder-only · TSP100R^2 of fit: R^2 0.794 vs 0.989 decoupledHurt
scale to D=42, W=512Decoder-only · TSPLib (all)optimality gap (%): 5.235 -> 1.288Improved

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