NLP Generative AI - Data Extraction & Generation with LLMs

About XA Group:

At XA Group, we are dedicated to driving substantial technological advancements in the automotive and insurance sectors. Our mission is to empower businesses with intelligent solutions, making them smarter, safer, and more efficient.

We are seeking an intern who can research, implement, present metrics, and demonstrate the following tasks related to (long context) data extraction and generation using Transformers & Large Language Models (LLMs):

Key Responsibilities:

• • 1. Instruction Fine-tuning for Documents (Text and Tables):
    • a. Fine-tune LLMs for document-specific instruction understanding, including text and tables.
    • b. Work on table detection (bordered and borderless), table structure detection, and mapping tables with text.
    • c. Implement a system for table RAG.
    • 2. Create Instruction Datasets with LLM Approaches:
    • a. Specialize in creating instructive datasets with long context text and tables using LLM approaches.
    • b. Develop expertise in generating specialized datasets tailored to the domain, focusing on specific instruction understanding.
    • 3. Fine-tune LLM Models for Q&A in Domain-specific Contexts:
    • a. Implement fine-tuning strategies for LLMs to extract information specific to the domain for Question and Answer (Q&A) tasks.
    • b. Showcase the model's ability to understand and respond to queries within the specified domain context.
    • 4. Model Quantization for Inference Speed and Accuracy:
    • a. Investigate model quantization methods to optimize LLMs for inference speed and accuracy, especially on GPUs.
    • b. Provide benchmarks and metrics for different quantization approaches, emphasizing trade-offs between speed and accuracy.
    • 5. Model Evaluation and Metrics:
    • a. Develop comprehensive evaluation metrics for LLM performance in data extraction and generation tasks.
    • b. Present findings through clear and concise reports, including visualizations and comparisons.

Requirements:

• • · Background in Generative AI, Natural Language Processing (NLP) and machine learning.
    • · Proficiency in programming languages such as Python and familiarity with relevant libraries (e.g., TensorFlow, PyTorch). Worked with LLMs, hugging face transformers.
    • · Strong analytical and research skills.
    • · Effective communication skills, including the ability to present findings to stakeholders.
    • · Ability to work independently and as part of a team

Perks:

• • Mentorship from industry experts in the field of Computer Vision.
    • Hands-on experience with cutting-edge technologies and real-world applications.
    • Opportunity to contribute to projects with meaningful impact.
    • Collaborative and innovative work environment.

$800 - $1,000 a month

Apply for this job