Model Optimization | Mobina Kashaniyan

LLM-Driven AutoML and AI Agents for Multilingual Handwritten OCR

Thu, 02 Jul 2026 00:00:00 +0000

LLM-Driven AutoML for Handwritten Text Recognition

Handwritten text recognition (HTR), also known as handwritten OCR, remains a challenging problem in computer vision and document AI, particularly across different languages and writing systems.

Arabic, Persian, and English handwriting differ substantially in character structure, stroke patterns, connectivity, and visual appearance. These differences make multilingual and cross-lingual OCR difficult to address with manually designed neural network architectures.

In our paper, “LLM-Driven AutoML for Cross-Lingual Handwritten OCR: Closed-Loop Neural Architecture Search with GPT-5, GPT-4o, and Claude Sonnet 4,” we investigate whether large language models can act as autonomous AutoML agents for designing handwritten text recognition systems.

Large Language Models as AutoML Agents

Instead of manually selecting neural network architectures, preprocessing strategies, and model configurations, our framework uses large language models to propose, evaluate, and iteratively refine OCR architectures.

GPT-5, GPT-4o, and Claude Sonnet 4 independently participate in a closed-loop neural architecture search process, allowing the system to explore and improve candidate architectures with limited human intervention.

This connects several active research areas:

large language model agents
automated machine learning
neural architecture search
computer vision
document intelligence
handwritten text recognition
multilingual AI

Multilingual and Cross-Lingual OCR

The framework is evaluated on Arabic, Persian, and English handwritten text recognition, making the work relevant to multilingual OCR and low-resource language technologies.

Arabic and Persian handwriting recognition are particularly important because many OCR systems and datasets remain concentrated on English and other high-resource languages.

The proposed approach therefore explores whether LLM-driven automated model design can generalize across substantially different writing systems.

Closed-Loop Neural Architecture Search

Traditional neural architecture search can require substantial manual engineering or large computational search spaces.

Our approach investigates an alternative in which LLM agents reason about experimental results and iteratively propose new neural architectures.

This creates a closed-loop process:

architecture generation → model training → evaluation → performance feedback → architecture refinement

The goal is to automate more of the model-development process while maintaining interpretable experimental iterations.

Applications in Document AI

The work is relevant beyond handwritten OCR. Similar LLM-driven AutoML approaches could potentially support:

document understanding
historical document recognition
multilingual document processing
digitization of handwritten archives
low-resource language technologies
automated computer vision model design
intelligent document processing

This research contributes to LLM-driven AutoML, handwritten text recognition, multilingual OCR, neural architecture search, document AI, LLM agents, low-resource language processing, and automated neural network design.

AI Agents for Automated Machine Learning

A central idea of this work is to use large language models as AI agents for AutoML. Rather than treating an LLM only as a text generator, the framework gives the model an iterative role in neural architecture design.

Each LLM agent can propose an OCR architecture, observe training and evaluation results, reason about model performance, and generate a refined architecture for the next iteration.

This creates an agentic AI workflow:

AI agent → architecture proposal → training → evaluation → feedback → architecture refinement

The framework therefore connects AI agents, LLM agents, agentic AI, AutoML, neural architecture search, and multilingual handwritten OCR in a closed-loop model-design process.

Paper

LLM-Driven AutoML for Cross-Lingual Handwritten OCR: Closed-Loop Neural Architecture Search with GPT-5, GPT-4o, and Claude Sonnet 4

Mobina Kashaniyan, Amirhossein Ghassemi, Nasser Mozayani

Published in 2025 15th International Conference on Computer and Knowledge Engineering (ICCKE), IEEE.

DOI: https://doi.org/10.1109/ICCKE68588.2025.11273810

PDF: https://mobinakashaniyan.github.io/papers/llm-driven-automl-cross-lingual-handwritten-ocr.pdf

IEEE Xplore: https://ieeexplore.ieee.org/abstract/document/11273810

Keywords

handwritten text recognition, HTR, handwritten OCR, optical character recognition, multilingual OCR, Persian OCR, Arabic OCR, Arabic handwriting recognition, Persian handwriting recognition, LLM-driven AutoML, neural architecture search, AutoML, document AI, document intelligence, LLM agents, multilingual AI, low-resource languages.

PerfMamba: Mamba Performance, SSM Pruning, Efficient Inference, and Model Optimization

Thu, 02 Jul 2026 00:00:00 +0000

Mamba Performance and Efficient State Space Models

Mamba and selective state space models (SSMs) have emerged as promising alternatives to Transformer-based sequence models for efficient sequence modeling and long-context processing.

Unlike conventional Transformer architectures that rely heavily on attention, Mamba-style state space models are designed to provide efficient sequence processing and favorable scaling behavior. However, practical deployment still depends on understanding runtime performance, memory usage, GPU utilization, inference efficiency, and computational bottlenecks.

Our paper, “PerfMamba: Performance Analysis and Pruning of Selective State Space Models,” provides an empirical performance analysis of Mamba-1 and Mamba-2 and investigates how state pruning can improve the efficiency of selective state space models.

Performance Analysis of Mamba-1 and Mamba-2

PerfMamba benchmarks Mamba-1 and Mamba-2 across sequence lengths ranging from 64 to 16,384 tokens.

The study analyzes:

runtime performance
execution time
computation patterns
memory access behavior
memory efficiency
I/O characteristics
GPU resource utilization
sequence-length scaling
component-level performance
model throughput

This analysis helps reveal how Mamba architectures behave in practice and where computational bottlenecks occur as sequence length increases.

Mamba Inference and Runtime Profiling

The theoretical efficiency of a neural architecture does not necessarily translate directly into optimal hardware performance.

PerfMamba therefore studies Mamba inference performance at the systems level, examining how different components of the architecture contribute to runtime, memory consumption, and resource utilization.

The profiling results show that the selective state space model component accounts for an important portion of the computational workload, making it a natural target for optimization.

This makes the work relevant to researchers studying:

Mamba inference
LLM inference optimization
AI inference acceleration
efficient deep learning
GPU performance
hardware-aware machine learning
machine learning systems

State Space Model Pruning

Based on the performance profiling results, PerfMamba investigates state pruning for selective state space models.

The proposed approach identifies and removes low-activity states within the SSM component to reduce unnecessary computation.

This connects Mamba optimization with broader research on:

neural network pruning
model compression
state-space model pruning
model acceleration
memory optimization
efficient AI inference
hardware-aware optimization

The goal is to improve computational efficiency while preserving model accuracy under moderate pruning.

Mamba Pruning for Faster and More Memory-Efficient Inference

PerfMamba demonstrates that selective state pruning can improve runtime and memory efficiency.

Under moderate pruning, the method achieves:

up to 1.14× inference speedup
up to 11.5% reduction in memory usage
improved throughput across different sequence lengths
limited degradation in model quality under moderate pruning

These results show that internal state redundancy can provide an opportunity for practical optimization of Mamba-based models.

Efficient LLM Inference and Mamba

State space models such as Mamba are increasingly relevant to research on large language models and efficient LLM inference.

As language models grow in size and context length, researchers are actively exploring architectures that can process long sequences with lower computational and memory overhead.

PerfMamba contributes to this area by examining how the internal components of Mamba behave at runtime and how selective pruning can reduce inference cost.

The work is therefore relevant to:

LLM efficiency
efficient LLM inference
long-context language models
inference optimization
sequence model acceleration
memory-efficient AI
scalable machine learning systems

Mamba vs Transformer Efficiency

Mamba and other state space models are often studied as alternatives to Transformer architectures because of their sequence-processing efficiency.

PerfMamba complements architectural comparisons by asking a systems-oriented question:

Where is computation actually spent inside Mamba, and which components can be optimized to improve real-world performance?

This performance-oriented perspective is important because theoretical complexity alone does not determine actual GPU runtime, memory behavior, or inference throughput.

Long-Sequence Modeling and Scaling

One of the key motivations for state space models is their ability to process long sequences efficiently.

PerfMamba evaluates Mamba models over sequence lengths from 64 to 16,384 tokens, providing insight into how execution time, memory use, and resource utilization change as workloads scale.

This makes the study relevant to:

long-sequence modeling
long-context models
scalable sequence models
efficient sequence processing
large language model systems
high-performance AI

Hardware-Aware Mamba Optimization

PerfMamba takes a systems-oriented approach to model optimization.

Rather than evaluating only predictive accuracy, the work considers:

computation
memory access
I/O
resource utilization
execution time
throughput
sequence-length scaling

This makes the work relevant to hardware-aware AI, GPU optimization, machine learning systems, high-performance computing, and AI systems research.

What Is PerfMamba?

PerfMamba is a performance-analysis and optimization study of Mamba-1 and Mamba-2 selective state space models.

The framework combines:

Mamba benchmarking → runtime profiling → bottleneck analysis → state pruning → inference acceleration → memory reduction

The goal is to understand how Mamba models behave on real hardware and identify opportunities for more efficient inference.

Why Does Mamba Pruning Matter?

Modern deep learning models often contain redundant computation.

For selective state space models, some internal states may contribute less to model behavior than others. Identifying and pruning low-activity states can potentially reduce computation and memory requirements.

PerfMamba demonstrates that this strategy can improve the efficiency of Mamba models while maintaining competitive model quality under moderate pruning.

Applications

The findings are relevant to researchers and engineers working on:

Mamba architectures
state space models
large language models
efficient LLM inference
long-context AI
model compression
neural network pruning
AI inference acceleration
GPU optimization
scalable machine learning
machine learning systems
high-performance computing
hardware-aware AI

Paper

PerfMamba: Performance Analysis and Pruning of Selective State Space Models

Abdullah Al Asif, Mobina Kashaniyan, Sixing Yu, Juan Pablo Muñoz, Ali Jannesari

Published in International Symposium on Benchmarking, Measuring and Optimization
Springer, Singapore, 2026
Pages 27–44

DOI:
https://doi.org/10.1007/978-981-95-9694-2_3

arXiv:
https://arxiv.org/abs/2511.22849

PDF:
https://mobinakashaniyan.github.io/papers/perfmamba-performance-analysis-pruning-selective-state-space-models.pdf

Keywords

PerfMamba, Mamba, Mamba architecture, Mamba model, Mamba-1, Mamba-2, Mamba performance, Mamba benchmarking, Mamba inference, Mamba optimization, Mamba pruning, Mamba efficiency, Mamba SSM, selective state space models, state space models, state-space models, SSM, SSM pruning, SSM optimization, SSM inference, efficient state space models, sequence models, sequence modeling, long sequence modeling, long-context models, Transformer alternatives, Mamba vs Transformer, large language models, LLM efficiency, efficient LLM inference, LLM inference optimization, AI inference, inference optimization, inference acceleration, model acceleration, model optimization, model pruning, neural network pruning, state pruning, model compression, neural network compression, efficient AI, efficient machine learning, efficient deep learning, model efficiency, memory efficiency, memory optimization, runtime performance, runtime profiling, performance analysis, performance optimization, AI benchmarking, GPU performance, GPU benchmarking, GPU optimization, hardware-aware AI, hardware-aware optimization, resource utilization, memory access patterns, I/O performance, sequence-length scaling, AI systems, machine learning systems, deep learning systems, high-performance computing, HPC, scalable machine learning.

Citation

Al Asif, A., Kashaniyan, M., Yu, S., Muñoz, J. P., & Jannesari, A. (2026). PerfMamba: Performance Analysis and Pruning of Selective State Space Models. In International Symposium on Benchmarking, Measuring and Optimization (pp. 27–44). Springer, Singapore. https://doi.org/10.1007/978-981-95-9694-2_3