@inproceedings{wu-2026-peft,
title = "{PEFT} Gains Need Hidden Baselines: Rank and Optimizer Sweeps Change {L}o{RA} Conclusions",
author = "Wu, Ding",
editor = "Sedoc, Jo{\~a}o and
Shu, Raphael and
Drozd, Aleksandr and
Akula, Arjun and
Rogers, Anna and
Giorgi, Salvatore and
Benton, Adrian",
booktitle = "Proceedings of the Seventh Workshop on Insights from Negative Results in {NLP}",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.insights-1.17/",
pages = "127--132",
ISBN = "979-8-89176-480-4",
abstract = "New parameter-efficient fine-tuning (PEFT) methods are often credited with gains against a single default LoRA configuration, even though inexpensive tuning of vanilla LoRA may change the comparison. We present a deliberately small, descriptive case study of two claims: adaptive rank allocation with AdaLoRA and LoRA-specific optimization with LoRA-RITE. Using RoBERTa-base on capped subsets of three GLUE classification tasks, we compare each method with a pre-specified vanilla LoRA control sweep. AdaLoRA beats fixed LoRA at rank 8 in 4 of 7 dataset-seed cells, but beats the best tested fixed rank in only 1 of 7. LoRA-RITE beats default AdamW in 3 of 6 cells, but tuned AdamW in only 1 of 6. Several SST-2 runs collapse to the majority label, so these counts are illustrative rather than population-level estimates. The practical contribution is an attribution protocol: report modest vanilla rank/learning-rate sweeps, per-cell margins, instability checks, and budget accounting before attributing gains to a specialized PEFT mechanism."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wu-2026-peft">
<titleInfo>
<title>PEFT Gains Need Hidden Baselines: Rank and Optimizer Sweeps Change LoRA Conclusions</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ding</namePart>
<namePart type="family">Wu</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 Seventh Workshop on Insights from Negative Results in NLP</title>
</titleInfo>
<name type="personal">
<namePart type="given">João</namePart>
<namePart type="family">Sedoc</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Raphael</namePart>
<namePart type="family">Shu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aleksandr</namePart>
<namePart type="family">Drozd</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Arjun</namePart>
<namePart type="family">Akula</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rogers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Salvatore</namePart>
<namePart type="family">Giorgi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Adrian</namePart>
<namePart type="family">Benton</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">979-8-89176-480-4</identifier>
</relatedItem>
<abstract>New parameter-efficient fine-tuning (PEFT) methods are often credited with gains against a single default LoRA configuration, even though inexpensive tuning of vanilla LoRA may change the comparison. We present a deliberately small, descriptive case study of two claims: adaptive rank allocation with AdaLoRA and LoRA-specific optimization with LoRA-RITE. Using RoBERTa-base on capped subsets of three GLUE classification tasks, we compare each method with a pre-specified vanilla LoRA control sweep. AdaLoRA beats fixed LoRA at rank 8 in 4 of 7 dataset-seed cells, but beats the best tested fixed rank in only 1 of 7. LoRA-RITE beats default AdamW in 3 of 6 cells, but tuned AdamW in only 1 of 6. Several SST-2 runs collapse to the majority label, so these counts are illustrative rather than population-level estimates. The practical contribution is an attribution protocol: report modest vanilla rank/learning-rate sweeps, per-cell margins, instability checks, and budget accounting before attributing gains to a specialized PEFT mechanism.</abstract>
<identifier type="citekey">wu-2026-peft</identifier>
<location>
<url>https://aclanthology.org/2026.insights-1.17/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>127</start>
<end>132</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T PEFT Gains Need Hidden Baselines: Rank and Optimizer Sweeps Change LoRA Conclusions
%A Wu, Ding
%Y Sedoc, João
%Y Shu, Raphael
%Y Drozd, Aleksandr
%Y Akula, Arjun
%Y Rogers, Anna
%Y Giorgi, Salvatore
%Y Benton, Adrian
%S Proceedings of the Seventh Workshop on Insights from Negative Results in NLP
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-480-4
%F wu-2026-peft
%X New parameter-efficient fine-tuning (PEFT) methods are often credited with gains against a single default LoRA configuration, even though inexpensive tuning of vanilla LoRA may change the comparison. We present a deliberately small, descriptive case study of two claims: adaptive rank allocation with AdaLoRA and LoRA-specific optimization with LoRA-RITE. Using RoBERTa-base on capped subsets of three GLUE classification tasks, we compare each method with a pre-specified vanilla LoRA control sweep. AdaLoRA beats fixed LoRA at rank 8 in 4 of 7 dataset-seed cells, but beats the best tested fixed rank in only 1 of 7. LoRA-RITE beats default AdamW in 3 of 6 cells, but tuned AdamW in only 1 of 6. Several SST-2 runs collapse to the majority label, so these counts are illustrative rather than population-level estimates. The practical contribution is an attribution protocol: report modest vanilla rank/learning-rate sweeps, per-cell margins, instability checks, and budget accounting before attributing gains to a specialized PEFT mechanism.
%U https://aclanthology.org/2026.insights-1.17/
%P 127-132
Markdown (Informal)
[PEFT Gains Need Hidden Baselines: Rank and Optimizer Sweeps Change LoRA Conclusions](https://aclanthology.org/2026.insights-1.17/) (Wu, insights 2026)
ACL