@inproceedings{prottasha-etal-2026-monkey,
title = "Monkey Jump: {M}o{E}-Style {PEFT} for Efficient Multi-Task Learning",
author = "Prottasha, Nusrat Jahan and
Kowsher, Md and
Yu, Chun-Nam and
Chen, Chen and
Garibay, Ozlem",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.801/",
pages = "18631--18674",
ISBN = "979-8-89176-459-0",
abstract = "Mixture-of-experts variants of parameter-efficient fine-tuning enable per-token specialization, but they introduce additional trainable routers and expert parameters, increasing memory and training costs. This undermines the core goal of parameter efficient finetuning. We propose Monkey Jump, which brings MoE-style specialization to PEFT without adding extra trainable parameters for experts and routers. Instead of introducing new PEFT adapters as experts, Monkey Jump treats the PEFT adapters already present in each Transformer block (e.g., query, key, value, up, and down projections) as implicit experts and routes tokens among them. Routing is performed via k-means clustering with EMA-updated centers{---}no gradients, no learned parameters. We theoretically show that token-wise routing increases expressivity and can outperform shared adapters by avoiding cancellation effects. In multi-task experiments spanning 14 text, 14 image, and 19 video benchmarks, Monkey Jump achieves competitive performance with MoE-PEFT methods while using 5.3{--}9.2{\texttimes} fewer trainable parameters than the MoE variant of the same base method, up to 48{\%} lower memory, and 1.5{--}2{\texttimes} faster training. Monkey Jump is architecture-agnostic and can be applied to any adapter-based PEFT method."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="prottasha-etal-2026-monkey">
<titleInfo>
<title>Monkey Jump: MoE-Style PEFT for Efficient Multi-Task Learning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nusrat</namePart>
<namePart type="given">Jahan</namePart>
<namePart type="family">Prottasha</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Md</namePart>
<namePart type="family">Kowsher</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chun-Nam</namePart>
<namePart type="family">Yu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chen</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ozlem</namePart>
<namePart type="family">Garibay</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Mixture-of-experts variants of parameter-efficient fine-tuning enable per-token specialization, but they introduce additional trainable routers and expert parameters, increasing memory and training costs. This undermines the core goal of parameter efficient finetuning. We propose Monkey Jump, which brings MoE-style specialization to PEFT without adding extra trainable parameters for experts and routers. Instead of introducing new PEFT adapters as experts, Monkey Jump treats the PEFT adapters already present in each Transformer block (e.g., query, key, value, up, and down projections) as implicit experts and routes tokens among them. Routing is performed via k-means clustering with EMA-updated centers—no gradients, no learned parameters. We theoretically show that token-wise routing increases expressivity and can outperform shared adapters by avoiding cancellation effects. In multi-task experiments spanning 14 text, 14 image, and 19 video benchmarks, Monkey Jump achieves competitive performance with MoE-PEFT methods while using 5.3–9.2× fewer trainable parameters than the MoE variant of the same base method, up to 48% lower memory, and 1.5–2× faster training. Monkey Jump is architecture-agnostic and can be applied to any adapter-based PEFT method.</abstract>
<identifier type="citekey">prottasha-etal-2026-monkey</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.801/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>18631</start>
<end>18674</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Monkey Jump: MoE-Style PEFT for Efficient Multi-Task Learning
%A Prottasha, Nusrat Jahan
%A Kowsher, Md
%A Yu, Chun-Nam
%A Chen, Chen
%A Garibay, Ozlem
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F prottasha-etal-2026-monkey
%X Mixture-of-experts variants of parameter-efficient fine-tuning enable per-token specialization, but they introduce additional trainable routers and expert parameters, increasing memory and training costs. This undermines the core goal of parameter efficient finetuning. We propose Monkey Jump, which brings MoE-style specialization to PEFT without adding extra trainable parameters for experts and routers. Instead of introducing new PEFT adapters as experts, Monkey Jump treats the PEFT adapters already present in each Transformer block (e.g., query, key, value, up, and down projections) as implicit experts and routes tokens among them. Routing is performed via k-means clustering with EMA-updated centers—no gradients, no learned parameters. We theoretically show that token-wise routing increases expressivity and can outperform shared adapters by avoiding cancellation effects. In multi-task experiments spanning 14 text, 14 image, and 19 video benchmarks, Monkey Jump achieves competitive performance with MoE-PEFT methods while using 5.3–9.2× fewer trainable parameters than the MoE variant of the same base method, up to 48% lower memory, and 1.5–2× faster training. Monkey Jump is architecture-agnostic and can be applied to any adapter-based PEFT method.
%U https://aclanthology.org/2026.emnlp-main.801/
%P 18631-18674
Markdown (Informal)
[Monkey Jump: MoE-Style PEFT for Efficient Multi-Task Learning](https://aclanthology.org/2026.emnlp-main.801/) (Prottasha et al., EMNLP 2026)
ACL
- Nusrat Jahan Prottasha, Md Kowsher, Chun-Nam Yu, Chen Chen, and Ozlem Garibay. 2026. Monkey Jump: MoE-Style PEFT for Efficient Multi-Task Learning. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 18631–18674, Budapest, Hungary. Association for Computational Linguistics.