arXiv subjects

Vasanth Iyer

Publications and source records attributed to Vasanth Iyer.

3 recordsLinked to original sources

arXiv2026

Dual-Node NVIDIA DGX Spark over Tailscale: A Remote-Access Testbed for Distributed LLM Training and Cyber-Threat-Intelligence Fine-Tuning

Compact AI systems make local language-model experimentation increasingly accessible, yet practical evidence for multi-node training on desktop-class accelerators remains limited. This report presents a proof-of-concept deployment of distributed NanoChat pretraining across two NVIDIA DGX Spark systems, each with a GB10 Grace Blackwell system-on-chip and 128 GB of unified memory, administered remotely over a Tailscale mesh VPN and connected for training by a dedicated 200 Gb/s QSFP56 direct fiber link. PyTorch torchrun, DDP, and NCCL were configured with one process per node, a depth-20 NanoChat model, a local batch size of 32 per node, and a 2,048-token context, giving a global batch of 131,072 tokens per step. The run sustained a step time of about 69.4 s (about 1,890 tokens/s), processing about 653 million tokens over four days. We document link configuration, container setup, interface binding, a step-zero evaluation bug that triggered NCCL timeouts, checkpointing, and troubleshooting lessons, as a reproducibility reference for small labs. We also built a cybersecurity fine-tuning dataset from 77 CISA advisories (338 training, 37 validation conversations) and ran a 17-question held-out evaluation comparing a baseline SFT checkpoint against a CTI-augmented checkpoint with an Ollama-hosted LLM judge. CTI-specific categories improved while general-knowledge categories regressed, for a small overall change from 2.06 to 2.29 on a 0-10 scale. The same cluster supports a 400-level AI course (CS 426) and a query engine for CompTIA Security+ POGIL activities in CBS 255, showing modest local infrastructure can serve both research and teaching. The study establishes feasibility rather than a scaling-efficiency claim, since single-node throughput used for comparison was estimated, not measured under matched conditions. Runbook and scripts are available (see Code Availability).

cs.AR↗

arXiv2026

Quantum-Enhanced Distributed Sensor Fusion: Lower Bounds on Aggregation from Projection Noise to Heisenberg-Limited Byzantine-Tolerant Networks

We derive unified lower bounds on the mean squared error (MSE) of distributed quantum sensor fusion under Byzantine faults and decoherence. Building on the classical Brooks-Iyengar overlap function and its vector extension, the predictive outlier model for virtual sensor tracking, and SPOTLESS spatial-temporal verification, we establish a two-parameter family of bounds indexed by entanglement visibility V and fault fraction f/M. For M quantum sensors with N atoms each and sensitivity eta, the MSE of any estimator satisfies MSE >= (1-V^2)/(4*N*eta^2*M_eff) + V^2/(4*N*eta^2*M_eff^2), where M_eff = M-2f under Brooks-Iyengar Byzantine fault tolerance and M_eff = M-f when predictive outlier detection successfully identifies faulty sensors. The bound interpolates continuously between the standard quantum limit (V=0, scaling as 1/sqrt(M_eff)) and the Heisenberg limit (V=1, scaling as 1/M_eff). Monte Carlo simulations with up to 64 sensors validate the theoretical scaling laws. Validation on the Intel Berkeley Lab 54-mote dataset with spatial clustering demonstrates 20-27 dB SNR improvement from entanglement per cluster, and reveals that missing classical sensor data degrades fusion agreement in the same pattern as quantum decoherence. The framework bridges quantum metrology with classical stream-processing architectures including Data-Cleaning Trees and the 80-20 Power Law for scale-invariant clustering.

cs.DC↗

arXiv2026

CyberLLM-FINDS 2025: Instruction-Tuned Fine-tuning of Domain-Specific LLMs with Retrieval-Augmented Generation and Graph Integration for MITRE Evaluation

Large Language Models (LLMs) such as Gemma-2B have shown strong performance in various natural language processing tasks. However, general-purpose models often lack the domain expertise required for cybersecurity applications. This work presents a methodology to fine-tune the Gemma-2B model into a domain-specific cybersecurity LLM. We detail the processes of dataset preparation, fine-tuning, and synthetic data generation, along with implications for real-world applications in threat detection, forensic investigation, and attack analysis. Experiments highlight challenges in prompt length distribution during domain-specific fine-tuning. Uneven prompt lengths limit the model's effective use of the context window, constraining local inference to 200-400 tokens despite hardware support for longer sequences. Chain-of-thought styled prompts, paired with quantized weights, yielded the best performance under these constraints. To address context limitations, we employed a hybrid strategy using cloud LLMs for synthetic data generation and local fine-tuning for deployment efficiency. To extend the evaluation, we introduce a Retrieval-Augmented Generation (RAG) pipeline and graph-based reasoning framework. This approach enables structured alignment with MITRE ATT&CK techniques through STIX-based threat intelligence, enhancing recall in multi-hop and long-context scenarios. Graph modules encode entity-neighborhood context and tactic chains, helping mitigate the constraints of short prompt windows. Results demonstrate improved model alignment with tactic, technique, and procedure (TTP) coverage, validating the utility of graph-augmented LLMs in cybersecurity threat intelligence applications.

cs.CR↗