@inproceedings{rafiuddin-abdullah-2026-context,
title = "Context-Conditioned Masked {L}o{RA}: Dynamic Rank Routing for Compute-Efficient Parameter-Efficient Fine-Tuning",
author = "Rafiuddin, Rifat and
Abdullah, Rafae",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-acl.1329/",
pages = "26670--26689",
ISBN = "979-8-89176-395-1",
abstract = "Parameter-efficient fine-tuning methods such as LoRA reduce trainable parameters, but still apply dense low-rank updates per token, leaving adaptation compute largely fixed once rank is set. We propose Context-Conditioned Masked LoRA (CCM-LoRA), which learns a lightweight router that activates an input-dependent subset of LoRA rank directions, turning LoRA into dynamic rank routing and enabling contextual sparsity in fine-tuning and inference. CCM-LoRA is trained with a budget-constrained objective that targets an expected effective rank (or FLOPs) while regularizing routing to avoid degenerate always-on/off masks. Across public NLU and multilingual benchmarks, CCM-LoRA improves the accuracy{--}efficiency Pareto frontier versus static-rank LoRA and adaptive-rank baselines, matching or improving task performance at lower inference-time effective rank. We also provide a reproducible profiling protocol and analyses of rank usage, router overhead, and robustness under domain and language shift."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="rafiuddin-abdullah-2026-context">
<titleInfo>
<title>Context-Conditioned Masked LoRA: Dynamic Rank Routing for Compute-Efficient Parameter-Efficient Fine-Tuning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Rifat</namePart>
<namePart type="family">Rafiuddin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rafae</namePart>
<namePart type="family">Abdullah</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-395-1</identifier>
</relatedItem>
<abstract>Parameter-efficient fine-tuning methods such as LoRA reduce trainable parameters, but still apply dense low-rank updates per token, leaving adaptation compute largely fixed once rank is set. We propose Context-Conditioned Masked LoRA (CCM-LoRA), which learns a lightweight router that activates an input-dependent subset of LoRA rank directions, turning LoRA into dynamic rank routing and enabling contextual sparsity in fine-tuning and inference. CCM-LoRA is trained with a budget-constrained objective that targets an expected effective rank (or FLOPs) while regularizing routing to avoid degenerate always-on/off masks. Across public NLU and multilingual benchmarks, CCM-LoRA improves the accuracy–efficiency Pareto frontier versus static-rank LoRA and adaptive-rank baselines, matching or improving task performance at lower inference-time effective rank. We also provide a reproducible profiling protocol and analyses of rank usage, router overhead, and robustness under domain and language shift.</abstract>
<identifier type="citekey">rafiuddin-abdullah-2026-context</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.1329/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>26670</start>
<end>26689</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Context-Conditioned Masked LoRA: Dynamic Rank Routing for Compute-Efficient Parameter-Efficient Fine-Tuning
%A Rafiuddin, Rifat
%A Abdullah, Rafae
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F rafiuddin-abdullah-2026-context
%X Parameter-efficient fine-tuning methods such as LoRA reduce trainable parameters, but still apply dense low-rank updates per token, leaving adaptation compute largely fixed once rank is set. We propose Context-Conditioned Masked LoRA (CCM-LoRA), which learns a lightweight router that activates an input-dependent subset of LoRA rank directions, turning LoRA into dynamic rank routing and enabling contextual sparsity in fine-tuning and inference. CCM-LoRA is trained with a budget-constrained objective that targets an expected effective rank (or FLOPs) while regularizing routing to avoid degenerate always-on/off masks. Across public NLU and multilingual benchmarks, CCM-LoRA improves the accuracy–efficiency Pareto frontier versus static-rank LoRA and adaptive-rank baselines, matching or improving task performance at lower inference-time effective rank. We also provide a reproducible profiling protocol and analyses of rank usage, router overhead, and robustness under domain and language shift.
%U https://aclanthology.org/2026.findings-acl.1329/
%P 26670-26689
Markdown (Informal)
[Context-Conditioned Masked LoRA: Dynamic Rank Routing for Compute-Efficient Parameter-Efficient Fine-Tuning](https://aclanthology.org/2026.findings-acl.1329/) (Rafiuddin & Abdullah, Findings 2026)
ACL