CSE · SEMINAR TOPIC Generative AI and Large Language Mod…
Computer Science Engineering Seminar Report

Generative AI and Large Language Models

Generative AI refers to systems that create new content — text, images, code, audio — by learning patterns from massive datasets.

Large Language Models (LLMs) like the GPT and Claude families are built on the transformer architecture and have transformed natural language understanding and generation.

The Transformer Architecture

Transformers use a self-attention mechanism that lets the model weigh the importance of every word relative to others in a sequence, capturing long-range context. This replaced older recurrent networks and enabled training on enormous datasets.

LLMs are pre-trained on vast text corpora to predict the next token, then fine-tuned and aligned using techniques such as reinforcement learning from human feedback.

Quick Facts

AspectDetails
BranchComputer Science Engineering (CSE)
Topic TypeTechnical Seminar / Project Report
DifficultyIntermediate – Advanced
Best ForFinal-year BTech seminars & presentations
IncludesExplanation, key points, FAQs & references

Important Points to Remember

  • Built on the transformer architecture using self-attention.
  • Trained on huge datasets to predict the next token.
  • Generate text, code, images, and audio.
  • Applications: chatbots, summarization, translation, coding assistants.
  • Concerns: hallucination, bias, misuse, and high compute cost.
  • Alignment techniques improve safety and helpfulness.

Frequently Asked Questions

Generative AI is artificial intelligence that produces new content such as text, images, or code by learning patterns from large training datasets.

A large language model is a neural network trained on massive amounts of text to understand and generate human-like language, typically using the transformer architecture.

Key risks include factual hallucinations, bias in outputs, potential misuse for misinformation, and significant energy and compute requirements.