arXiv subjects

Karthikeyan S

Publications and source records attributed to Karthikeyan S.

3 recordsLinked to original sources

arXiv2026

Lumina: An AI-Augmented Multiscale Material Informatics Framework for Extreme Aero-Chemo-Thermo-Mechanical Regimes

Predictive simulations and experimental design involving extreme aero-chemo-thermo-mechanical regimes require high-fidelity material representation across diverse physical states. However, data for metals, polymers, and propellants, explosives, and pyrotechnics (PEP) remain fragmented, obstructing traceability for formulators, experimentalists, and simulation engineers. This work introduces Lumina, a modular Python-based informatics framework that centralizes multiscale material data from atomistic simulation datasets to macro-scale experimental records, within a unified repository. Lumina employs a hierarchical XML-based schema and a dynamic runtime parsing mechanism to enable schema-independent parameter extraction. Beyond storage, the platform provides computational modules to visualize model fits, allowing experimentalists to optimize design of experiments (DoE) and formulators to validate chemical behaviors against benchmarks. This structured architecture serves as a high-fidelity pipeline for training machine learning models and enhancing the accuracy of predictive simulations. To streamline multi-disciplinary workflows, Lumina integrates a conversational AI assistant for intelligent material retrieval and natural language querying. By consolidating multiscale data into an extensible ecosystem, Lumina provides a scalable foundation for data-driven discovery and predictive modeling in advanced defense and aerospace engineering.

physics.comp-ph↗

arXiv2025

State-of-the-art Small Language Coder Model: Mify-Coder

We present Mify-Coder, a 2.5B-parameter code model trained on 4.2T tokens using a compute-optimal strategy built on the Mify-2.5B foundation model. Mify-Coder achieves comparable accuracy and safety while significantly outperforming much larger baseline models on standard coding and function-calling benchmarks, demonstrating that compact models can match frontier-grade models in code generation and agent-driven workflows. Our training pipeline combines high-quality curated sources with synthetic data generated through agentically designed prompts, refined iteratively using enterprise-grade evaluation datasets. LLM-based quality filtering further enhances data density, enabling frugal yet effective training. Through disciplined exploration of CPT-SFT objectives, data mixtures, and sampling dynamics, we deliver frontier-grade code intelligence within a single continuous training trajectory. Empirical evidence shows that principled data and compute discipline allow smaller models to achieve competitive accuracy, efficiency, and safety compliance. Quantized variants of Mify-Coder enable deployment on standard desktop environments without requiring specialized hardware.

cs.SE↗

arXiv2025

Efficient Knowledge Feeding to Language Models: A Novel Integrated Encoder-Decoder Architecture

This paper introduces a novel approach to efficiently feeding knowledge to language models (LLMs) during prediction by integrating retrieval and generation processes within a unified framework. While the Retrieval-Augmented Generation (RAG) model addresses gaps in LLMs' training data and knowledge limits, it is hindered by token limit restrictions and dependency on the retrieval system's accuracy. Our proposed architecture incorporates in-context vectors (ICV) to overcome these challenges. ICV recasts in-context learning by using latent embeddings of LLMs to create a vector that captures essential task information. This vector is then used to shift the latent states of the LLM, enhancing the generation process without adding demonstration examples to the prompt. ICV directly integrates information into the model, enabling it to process this information more effectively. Our extensive experimental evaluation demonstrates that ICV outperforms standard in-context learning and fine-tuning across question-answering, information retrieval, and other tasks. This approach mitigates the limitations of current RAG models and offers a more robust solution for handling extensive and diverse datasets. Despite leveraging a fraction of the parameters, our ICV-enhanced model achieves competitive performance against models like LLaMA-3, Gemma, and Phi-3, significantly reducing computational costs and memory requirements. ICV reduces prompt length, is easy to control, surpasses token limitations, and is computationally efficient compared to fine-tuning.

cs.CL↗