The 1st International Workshop · Sustainable & Efficient LVA Models

The 1st International Workshop on Sustainable and Efficient Language, Vision, and Action Models (SELVA)

Confronting the computational costs of scaling Language, Vision, and Action Models.

ACL 2026 Co-located Workshop San Diego, California, USA Tentative: one-day online workshop

Overview

SELVA focuses on efficiency and sustainability across language, vision, and action models, especially for real-time and resource-constrained deployment.

Recent advances in LLMs, VLMs, and VLA systems are accompanied by rapidly increasing compute and energy costs. SELVA aims to foster cross-disciplinary discussion spanning NLP, computer vision, robotics, and systems to develop methods that are both high-performing and computationally feasible.

Efficient Architectures for Multimodality
MoE, state-space models, and sparse routing for language-vision-action fusion.

Model Compression and Acceleration
Sub-4-bit quantization, structured pruning, and distillation.

Sustainable Training and Real-Time Inference
Energy-aware pretraining, PEFT, and optimized engines for low latency.

Efficiency for Embodied AI
Efficient world models, policy optimization, and low-compute interaction loops.

Metrics, Measurement, and Societal Impact
Transparent energy/latency reporting and broader environmental implications.

Call for Papers

We welcome high-quality original research and system papers related to efficient and sustainable LVA models.

Submission Scope

Efficient multimodal and embodied AI architectures
Compression, quantization, pruning, and distillation for foundation models
Systems optimization and low-latency deployment pipelines
Energy/latency benchmarks and trustworthy reporting
Sustainable AI policy and broader impact analyses

Hybrid format planned: invited talks, contributed talks, poster session, and award ceremony.

Submission Website

All submissions must be in PDF format, anonymized, and prepared using the ACL Article Template. This workshop follows the ACL formatting guidelines: long papers may use either the Research Track or Industry Track template, while abstracts and short papers should follow the same format with a limit of 4 pages (including figures, tables and references). Authors should also follow the ACL 2026 author instructions .

Important Dates

Paper submission deadline

May 25, 2026

Notification

Jun 10, 2026

Camera-ready deadline

Jun 15, 2026

Workshop date

July 03, 2026

Submission Website

Submit via OpenReview

Timezone convention: AoE (UTC−12).

Important: We will feature Best Paper Award, and certificates will be presented during the workshop.

Workshop Schedule

July 3, 2026

Time	Event	Speaker / Paper
09:00	Opening Remarks
09:00–09:20	Invited Talk 1	Ruihao Gong
09:20–09:40	Invited Talk 2	Michele Magno
09:40–10:00	Invited Talk 3	Priya Panda
10:00–10:10	Oral Paper 1	QuantEL: Quantization-Informed Activation Editing Localization for Hallucination Mitigation in Quantized LLMs
10:10–10:20	Oral Paper 2	Understanding and Selecting Calibration Data for LLM Quantization: From Sensitivity Analysis to Activation-Based Curation
10:20–10:30	Break
10:30–10:50	Invited Talk 4	Huanrui Yang
10:50–11:10	Invited Talk 5	Yukang Chen
11:10–11:20	Oral Paper 3	Linear Predictability of Attention Heads in Large Language Models
11:20–11:30	Oral Paper 4	TiRO: Fast Discrete-Diffusion VLA Inference via Teacher-Initialized Rollout
11:30	Awarding & Closing

Invited Speakers

Invited speakers for the SELVA Workshop.

Priyadarshini Panda

University of Southern California

Algorithm-hardware co-design and neuromorphic computing for ultra-low-power AI systems.

Michele Magno

ETH Zürich

TinyML, smart sensing, and energy-efficient IoT systems for low-power edge intelligence.

Yukang Chen

NVIDIA

LongLoRA / LongVILA and efficient long-context modeling for interactive agents.

Huanrui Yang

University of Arizona

Mathematical foundations of efficient deployment and adaptation for robust AI models.

Ruihao Gong

Beihang University / SenseTime Research

Model quantization, sparsity, and deployment systems for efficient large-model inference.

Organizers

Core organizers and contributors listed in the submission.

Haotong Qin

ETH Zürich

Homepage

Weilun Feng

Institute of Computing Technology, Chinese Academy of Sciences

Homepage

Yawei Li

ETH Zürich

Homepage

Zixiang Zhao

ETH Zürich

Homepage

Zechun Liu

Jian Cheng

Institute of Automation, Chinese Academy of Sciences

Homepage

Olivera Kotevska

Oak Ridge National Laboratory

Homepage

Accepted Papers

Accepted submissions to SELVA 2026, grouped by presentation format.

Accepted Oral Papers

QuantEL: Quantization-Informed Activation Editing Localization for Hallucination Mitigation in Quantized LLMs [Paper]
Tianbo Wang; Lingyan Meng; Yuqing Ma; Zhange Zhang; Aishan Liu; Jinyang Guo; Jian Yang; Xianglong Liu
Understanding and Selecting Calibration Data for LLM Quantization: From Sensitivity Analysis to Activation-Based Curation [Paper]
Maoxiong; Benben Hai; Anzheng Wang; Xiangdong Liu; Lingwei Meng; Tingting Wang; Shuli Zheng; Yixin Zhou; Haibiao Chen
Linear Predictability of Attention Heads in Large Language Models [Paper]
Khalid Shaikh; Asmit Kumar Singh; Rebecca Dsouza; Shikhar Shiromani
TiRO: Fast Discrete-Diffusion VLA Inference via Teacher-Initialized Rollout [Paper]
Yuze Lv; Yao Mu

Accepted Poster Papers

NOEM³A: a Neuro-symbolic Ontology-Enhanced Method for Multi-intent understanding in Mobile Agents [Paper]
Ioannis Tzachristas; Aifen Sui
Edge-Optimized Offline Voice Assistants for Low-Resource Languages Tokenization and Deployment Insights from Garo [Paper]
Badal Nyalang
OSCAR: Offline Spectral Covariance-Aware Rotation for 2-bit KV Cache Quantization [Paper]
Zhongzhu Zhou; Donglin Zhuang; Jisen Li; Ziyan Chen; Shuaiwen Leon Song; Ben Athiwaratkun; Xiaoxia Wu
From Frontier to Frugal: Evaluating Self-Evolution Frameworks with Small Language Models [Paper]
Shayan Ali Akbar; Jiaming Qu; Chen Ling; Madhu Gopinathan; Erwin Cornejo
Parameter-Efficient Fine-Tuning for Low-Resource Languages: A Systematic Multi-Task Evaluation on Hausa Language [Paper]
Abdulhamid Abubakar; Shamsuddeen Hassan Muhammad; Idris Abdulmumin
MMViR: A Multi-Modal and Multi-Granularity Representation for Long-range Video Understanding [Paper]
Zizhong Li; Haopeng Zhang; Jiawei Zhang
F-WANDA: Fisher-Reweighted Post-Training Pruning for Sustainable Deployment of Large Language Models [Paper]
Himanshu Mishra
Less is More: Label-Guided Efficient Summarization of Procedural Videos [Paper]
Shreya Rajpal; Michal Golovanevsky; Carsten Eickhoff
Do Knowledge Cutoffs Drive Clinical Accuracy and User Trust? Quantifying Temporal Decay in Large Language Models [Paper]
Michael Cacioli; Aryan Arya; Austen Liao; Kevin Zhu
Compact Metrics, Catastrophic Failures: Trustworthy Benchmarking of Vision-Language Foundation Models in 3D Medical Segmentation [Paper]
Miroslaw Narbutt; Tanmoy Debnath

Program Committee (Reviewers)

We thank the following reviewers for their service.

Abdulhamid Abubakar
Badal Nyalang
Yash Ganpat Sawant
Tanmoy Debnath
Yuze Lv
Jiakai Wang
Bogdan Bogachov
Yulong Cao
Yisong Xiao
Himanshu Mishra
Toprak Seda Karaosmanoğlu
Aryan Arya
Mingyuan Zhang
Keshav Gupta
Jun Guo
Awais Rauf
Xiaoxia Wu
Jiaming Qu
Ziyan Chen
Diksha Agarwal
Mohan Premchand Bhambhani
Madhu Gopinathan
Michael Cacioli
Yan Wang
Duorui Wang
Tianlin Li
Ruihao Gong
KOTTILINGAM KOTTURSAMY
Ioannis Tzachristas
Jisen Li
Jiachen Sun
Guoxin Fan
Donglin Zhuang
Sávio Salvarino Teles de Oliveira
Md. Shahriar Karim
Xiaokun Liu
Mingqiang Wu
Zizhong Li
Hang Yu
Zhongzhu Zhou

Shikhar Shiromani
Jan Dobrosolski
Tianbo Wang
Shayan Ali Akbar
Shreya Rajpal
Michal Golovanevsky
Ruikui Wang
cliu@cmri.org.au
Xiaowei Zhao
Yuqing Ma
Jingzhi Li
Marshiat Mithe Syed
Nikita Letov
Anzheng Wang
Yaoyao Fiona Zhao
Khalid Shaikh

Venue & Logistics

ACL 2026 is scheduled in San Diego, and workshop logistics can align with the ACL workshop venue details.

Co-located Conference

The 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026)

San Diego, California, USA · July 2–7, 2026

Workshop days: July 3–4, 2026 (SELVA exact date/room TBD).

Conference Venue (ACL 2026)

Grand Hyatt Manchester San Diego
1 Market Place, San Diego, CA 92101

ACL 2026 venue page · Hotel booking link