Learn How To Observe, Manage, and Scale, Agentic AI Apps Using Azure AI Foundry - with this hands-on workshop

[ICML 2025] Official code for the paper "RoSTE: An Efficient Quantization-Aware Supervised Fine-Tuning Approach for Large Language Models"

Code for SFT and RL

[ICML 2025] Official code for the paper "RoSTE: An Efficient Quantization-Aware Supervised Fine-Tuning Approach for Large Language Models"

LoRA fine-tuning pipeline for tool-calling chat LLMs with config-driven datasets, deterministic prompts, and built-in tool-call evaluation.

Automatic music tagging using foundation models

Fine-tuned Meta's LLaMA 3.2 1B for text summarization using QLoRA (4-bit quantization + LoRA), achieving 40%+ improvement in ROUGE-2 over the base model on CNN/DailyMail dataset.

🎯 Fine-tuning LLMs using LlamaFactory for financial intent understanding | Evaluating open-source models on OpenFinData benchmark | Full implementation with multiple models (Qwen2.5/ChatGLM3/Baichuan2/Llama3)

Fine-tune Qwen2.5-VL-7B with LoRA to predict human-rated emotion intensity (1–7) from images, with a ResNet18 regression baseline, full preprocessing/SFT pipeline, and evaluation (MAE/RMSE + bias analysis).

Fine-tuning Llama-3 8B using Unsloth & QLoRA to automate SME customer service logic with 99% accuracy.

🦙 Llama2-FineTuning: Fine-tune LLAMA 2 with Custom Datasets Using LoRA and QLoRA Techniques

Supervised Fine Tuning with QLoRA

Fine-tuning various Llama 3.1 family of models on the Mult-It dataset

End-to-end Supervised Fine-Tuning (SFT) pipeline for TinyLlama-1.1B-Chat, specialized in trademark similarity risk assessment using heuristic-labeled SFT data, CPU-only LoRA training, adapter validation, full-weight merge, GGUF export, quantization (Q4_K_M), and local inference deployment via llama.cpp.

Compact TensorFlow language model for Election Commission of India (ECI) domain pretraining and assistant-masked SFT.

Fine-tune Qwen3-0.6B for resume parsing using LoRA

LoRA fine-tuning pipeline for tool-calling chat LLMs with config-driven datasets, deterministic prompts, and built-in tool-call evaluation.

STaR Self-Taught Reasoner implementation on GSM8K — Zero-Shot CoT vs Vanilla SFT vs STaR with Llama 3.2-3B

A Multimodal AI medical assistant

Fine-tuned LLaMA 3 (8B) using Unsloth with 4-bit quantization and LoRA-based PEFT to enable memory-efficient, accelerated training. Conducted supervised fine-tuning on the Alpaca Cleaned dataset using FP16 precision, gradient checkpointing, and 8-bit AdamW optimization, achieving effective instruction tuning on limited GPU resources.