cs.CLNov 4, 2026

Matching the Training-Time Weight Rule in Inference-Time Expert Deactivation

Authors: Hao Cheng, Zhiyu Yang, Bo Ma, Hanzhang Jia

Organizations: Zhejiang Gongshang University · Chang'an University · University of Sussex

Abstract

Expert deactivation at inference time changes both token routing and expert weights, and its effect on performance depends on how these changes are implemented. We compare removal, freezing, remaining-expert renormalization, and unnormalized re-routing on identical disabled sets across five instruction-tuned checkpoints from four model families, with a base OLMoE checkpoint as an additional comparison. We develop a weight-sum model that relates the loss of expert contributions and changes in branch scale to the weight rule used during training. With half of the experts deactivated according to activation-based rankings, remaining-expert renormalization yields the highest perplexity in models trained with raw router probabilities, whereas unnormalized re-routing yields the highest perplexity in models trained with normalized weights. Perplexity ranges from 1.4 to 73 times the intact baseline when the training rule is retained, and from 14 to 28,740 times the baseline under a rule mismatch. Task outcomes vary across evaluation types: on OLMoE, GSM8K strict-match accuracy falls to zero under all four implementations. Extending the model to changes in the number of active experts yields a further prediction, supported by the OLMoE experiments: the perplexity penalty from additional renormalization increases as fewer experts are activated.

Figures & tables

Appendix figures & tables8 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

Sep 4, 2026

Training-Free Halving of Activated Experts in Fine-Grained Mixture-of-Experts Models

Modern fine-grained Mixture-of-Experts (MoE) models route each token to a small number of experts and renormalize their router probabilities. We show that this renormalization implicitly calibrates expert output gain to the training top-kk: reducing kk at inference changes not only which experts are used but also the strength of the expert branch. We separate these effects by activating the top k1k_1 experts while normalizing by the probability mass of the top k2k_2 experts, introducing one integer with no parameters, training, or measurable compute overhead. On Qwen3.6-35B-A3B, reducing from 8 to 4 experts causes a 4.65-point MMLU drop under standard renormalization but only 0.35 points with k2=16k_2=16, while halving routed-expert compute. The result replicates on the 11×11\times larger Qwen3.5-397B-A17B, where reducing from 10 to 5 experts loses only 0.55 points with an appropriate reference set. Removing renormalization entirely is catastrophic, showing that preserving a suitable reference mass is crucial. We further find that perplexity and downstream accuracy favor different k2k_2, cautioning against selecting MoE compression settings using unlabeled text alone. Analyses also show that expert identity matters substantially more than expert weighting, while balanced and domain-specialized routing leaves limited room for expert pruning.
Oct 30, 2026

What We Keep of What We Forget: Replacement Operators and Routing Semantics in Training-Free MoE Expert Pruning

Training-free expert pruning makes two decisions: which experts to remove, and what the layer computes when its router still selects a removed expert. Published methods change both at once. Holding the mask fixed, we compare replacement operators from three families on OLMoE, DeepSeek-V2-Lite and Moonlight, with Granite as a fourth router, using paired, multiplicity-corrected tests. The largest effect is a choice usually left implicit: the survivors' gate scale. On routers trained without top-k renormalization, renormalizing the survivors' gates, a common implementation choice, costs 0.8-8.5 points and reverses whether rerouting beats MoNE's mean-output replacement; on two routers trained with renormalization, the trained scale wins at 50% pruning. We introduce the tilted novice, a closed-form low-rank map of the pruned expert's input whose rank-0 case is MoNE's constant. At 50% pruning it adds 1.2-1.8% parameters and gains 3.45 and 1.37 points over the constant on OLMoE and Moonlight, stable across calibration draws and largely unchanged by 4- and 3-bit quantization. No operator dominates: trained-scale rerouting is best on DeepSeek-V2-Lite, and a partial copy of each pruned expert is statistically indistinguishable from the tilted novice on two models. Evaluation can change conclusions: at 25% pruning MoNE's constant keeps 92-97% of dense multiple-choice accuracy but only 5-65% of GSM8K and 0-17% of code scores, and the calibration domain moves the mask 4-9 times more than resampling. We propose a five-item reporting checklist.
Nov 11, 2026

Fewer Expert Transfers Without Changing Routing: Expert Eviction for MoE Offloading

A mixture-of-experts (MoE) language model uses only a few experts per token, so a server can keep most expert weights in CPU memory and copy an expert to the GPU when the router requests it. Recent methods train the router to keep the same experts for several tokens, and treat fewer routing switches as a proxy for fewer copies. We measure the trade-off directly, in a harness that performs and times real CPU-to-GPU copies for OLMoE-1B-7B. On an L40S GPU, every tested policy that holds an expert set fixed and decodes faster than the least-recently-used (LRU) cache increases next-token loss by at least 0.067 nats. A change that leaves every routing decision intact does better: when the cache must evict an expert, evict the one whose next use lies farthest in the future. This rule copies 2.1x fewer experts than LRU at no loss in quality and is within 3.5% of the fewest copies possible. It needs future requests; the next sixteen tokens of demand recover 96% of its reduction, whereas a rule based on counts of past requests recovers 27%.