@inproceedings{munde-etal-2026-structured,
title = "Structured Pruning and Lightweight Adaptation of Foundation Models for Resource-Efficient Earth Observation",
author = "Munde, Aniket Ramchandra and
Nareddy, Kartheek Kumar Reddy and
Niebling, Julia",
editor = "Wang, Yimu and
Choi, Yee Man and
Wu, Di and
Ouyang, Siqi and
Azadani, Mozhgan Nasr and
Fung, Yi R. (May)",
booktitle = "Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently ({G}round{LM} 2026)",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.groundlm-1.27/",
pages = "343--352",
ISBN = "979-8-89176-478-1",
abstract = "Structured pruning has gained attention as a promising solution for the cost-effective deployment of large language, vision-language, and domain-specific foundation models. However, most prior work has focused on general-purpose language models and often lacks a systematic comparison of pruning strategies across model scales, architectures, and real-world multimodal applications. In this paper, we conduct a comprehensive evaluation of structured pruning methods, focusing on depth pruning, width pruning, and post-pruning recovery with parameter-efficient fine-tuning. We evaluate multiple open-source models, including Qwen3-0.6B, Qwen3-4B-Instruct, Qwen3-8B, LLaVA-OneVision-0.5B, and Phi-4-Mini-Instruct, across intrinsic metrics and downstream benchmarks such as BBH, MATH Hard, MuSR, and MMLU-Pro. Our findings reveal that (1) depth pruning generally provides a more stable trade-off between efficiency and performance than width pruning; (2) width pruning can reduce memory usage, but often gives limited throughput gains and less predictable accuracy retention; (3) reasoning-intensive benchmarks, particularly mathematical reasoning tasks, are more sensitive to structured pruning than some other evaluation tasks; (4) LoRA-based post-pruning adaptation can substantially recover performance after moderate pruning; and (5) pruning behaviors observed in general-purpose models transfer to Earth observation foundation models, as demonstrated on TerraMind, where depth pruning reduces latency and GPU memory usage while maintaining competitive EuroSAT and Bigearthnet task accuracy."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="munde-etal-2026-structured">
<titleInfo>
<title>Structured Pruning and Lightweight Adaptation of Foundation Models for Resource-Efficient Earth Observation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Aniket</namePart>
<namePart type="given">Ramchandra</namePart>
<namePart type="family">Munde</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kartheek</namePart>
<namePart type="given">Kumar</namePart>
<namePart type="given">Reddy</namePart>
<namePart type="family">Nareddy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Julia</namePart>
<namePart type="family">Niebling</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 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently (GroundLM 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yimu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yee</namePart>
<namePart type="given">Man</namePart>
<namePart type="family">Choi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Di</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Siqi</namePart>
<namePart type="family">Ouyang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mozhgan</namePart>
<namePart type="given">Nasr</namePart>
<namePart type="family">Azadani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yi</namePart>
<namePart type="given">R</namePart>
<namePart type="given">(May)</namePart>
<namePart type="family">Fung</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">979-8-89176-478-1</identifier>
</relatedItem>
<abstract>Structured pruning has gained attention as a promising solution for the cost-effective deployment of large language, vision-language, and domain-specific foundation models. However, most prior work has focused on general-purpose language models and often lacks a systematic comparison of pruning strategies across model scales, architectures, and real-world multimodal applications. In this paper, we conduct a comprehensive evaluation of structured pruning methods, focusing on depth pruning, width pruning, and post-pruning recovery with parameter-efficient fine-tuning. We evaluate multiple open-source models, including Qwen3-0.6B, Qwen3-4B-Instruct, Qwen3-8B, LLaVA-OneVision-0.5B, and Phi-4-Mini-Instruct, across intrinsic metrics and downstream benchmarks such as BBH, MATH Hard, MuSR, and MMLU-Pro. Our findings reveal that (1) depth pruning generally provides a more stable trade-off between efficiency and performance than width pruning; (2) width pruning can reduce memory usage, but often gives limited throughput gains and less predictable accuracy retention; (3) reasoning-intensive benchmarks, particularly mathematical reasoning tasks, are more sensitive to structured pruning than some other evaluation tasks; (4) LoRA-based post-pruning adaptation can substantially recover performance after moderate pruning; and (5) pruning behaviors observed in general-purpose models transfer to Earth observation foundation models, as demonstrated on TerraMind, where depth pruning reduces latency and GPU memory usage while maintaining competitive EuroSAT and Bigearthnet task accuracy.</abstract>
<identifier type="citekey">munde-etal-2026-structured</identifier>
<location>
<url>https://aclanthology.org/2026.groundlm-1.27/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>343</start>
<end>352</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Structured Pruning and Lightweight Adaptation of Foundation Models for Resource-Efficient Earth Observation
%A Munde, Aniket Ramchandra
%A Nareddy, Kartheek Kumar Reddy
%A Niebling, Julia
%Y Wang, Yimu
%Y Choi, Yee Man
%Y Wu, Di
%Y Ouyang, Siqi
%Y Azadani, Mozhgan Nasr
%Y Fung, Yi R. (May)
%S Proceedings of the 1st Workshop on Grounding Language Models: Learning Faithfully and Efficiently (GroundLM 2026)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-478-1
%F munde-etal-2026-structured
%X Structured pruning has gained attention as a promising solution for the cost-effective deployment of large language, vision-language, and domain-specific foundation models. However, most prior work has focused on general-purpose language models and often lacks a systematic comparison of pruning strategies across model scales, architectures, and real-world multimodal applications. In this paper, we conduct a comprehensive evaluation of structured pruning methods, focusing on depth pruning, width pruning, and post-pruning recovery with parameter-efficient fine-tuning. We evaluate multiple open-source models, including Qwen3-0.6B, Qwen3-4B-Instruct, Qwen3-8B, LLaVA-OneVision-0.5B, and Phi-4-Mini-Instruct, across intrinsic metrics and downstream benchmarks such as BBH, MATH Hard, MuSR, and MMLU-Pro. Our findings reveal that (1) depth pruning generally provides a more stable trade-off between efficiency and performance than width pruning; (2) width pruning can reduce memory usage, but often gives limited throughput gains and less predictable accuracy retention; (3) reasoning-intensive benchmarks, particularly mathematical reasoning tasks, are more sensitive to structured pruning than some other evaluation tasks; (4) LoRA-based post-pruning adaptation can substantially recover performance after moderate pruning; and (5) pruning behaviors observed in general-purpose models transfer to Earth observation foundation models, as demonstrated on TerraMind, where depth pruning reduces latency and GPU memory usage while maintaining competitive EuroSAT and Bigearthnet task accuracy.
%U https://aclanthology.org/2026.groundlm-1.27/
%P 343-352
Markdown (Informal)
[Structured Pruning and Lightweight Adaptation of Foundation Models for Resource-Efficient Earth Observation](https://aclanthology.org/2026.groundlm-1.27/) (Munde et al., GroundLM 2026)
ACL