Changes between Version 2 and Version 3 of private/NlpInPracticeCourse/GenerativeLanguageModels

Timestamp:

Dec 10, 2024, 10:44:43 PM (7 months ago)

Author:

Ales Horak

Comment:

--

private/NlpInPracticeCourse/GenerativeLanguageModels

@@ -3 +3 @@
 [[https://is.muni.cz/auth/predmet/fi/ia161|IA161 NLP in Practice Course]], Course Guarantee: Aleš Horák
 
-Prepared by: Tomáš Foltýnek
+Prepared by: Aleš Horák
 
 == State of the Art ==
 
-Generating the text is - in principle - the same as predicting the following word. Given the seed (typically a start of the text), the model can predict/generate a following word that fits the context. Current state-of-the-art models are based on the Generative Pre-trained Transformer (GPT) architecture, which uses a multi-head attention mechanism to capture contextual features. The models contain several attention blocks to perform higher-order cognitive tasks.
+Generating text essentially involves predicting the next word in a sequence. Starting with an initial prompt (a seed), the model predicts or generates the subsequent word that aligns with the context. State-of-the-art models, like those based on the Generative Pre-trained Transformer (GPT) architecture,  are composed of multiple attention layers, enabling them to handle complex cognitive tasks.  
 
-The language models generate text regardless of factual correctness, which means that they may produce wrong, misleading or biased output. Some bias is deeply rooted in the training data, which are heavily unbalanced concerning genre and domain, as well as writers' gender, age and cultural background. In some applications, the bias may cause harmful outputs.
+Assistant models, such as ChatGPT, are built on this foundation, making them adept at understanding and generating human-like responses. A key aspect of using these models effectively is *prompt engineering*, which involves crafting well-structured inputs to guide the model's behavior and improve the quality of its output.
+
 
 === References ===
 
- 1. Vaswani, A. et al. (2017): Attention Is All You Need. ArXiv preprint: https://arxiv.org/abs/1706.03762
+ 1. Vaswani, A. et al. (2017): Attention Is All You Need. !ArXiv preprint: https://arxiv.org/abs/1706.03762
  1. Radford, A. et al. (2018): Improving Language Understanding by Generative Pre-Training. OpenAI: https://cdn.openai.com/research-covers/language-unsupervised/language_understanding_paper.pdf
  1. Navigli, R., Conia, S., & Ross, B. (2023). Biases in Large Language Models: Origins, Inventory, and Discussion. J. Data and Information Quality, 15(2). https://doi.org/10.1145/3597307
@@ -19 +20 @@
 == Practical Session ==
 
-We will be working with the [[https://colab.research.google.com/drive/19wZxHV6GLsRNvTdfVWbSK_vaoyEECHLj#scrollTo=PVXofXV4Ft7z|Google Colab Notebook]]. First, we load the GPT2-Large model and experiment with generating the text. To receive a more objective view of the probabilities of the following tokens, we adjust the generating function to give us the top k most words and their probabilities. Then, we learn how to calculate the perplexity of a given text.
-
-**Task 1: Exploring perplexity**
-
-Generate various text samples using different temperatures. Observe the relationship between temperature (parameter of the generator) and perplexity of the resulting text.
-
-**Task 1: Exploring bias**
-
-We will experiment with several prompts/seeds that are likely to produce biased output. 
-
-Your task will be to design more seeds and generate text or get predictions of subsequent words. Then, annotate the predictions (positive/negative/neutral), and answer the following questions:
-
-*  To which groups the GPT2 model outputs exhibit positive bias?
-*  To which groups the GPT2 model outputs exhibit negative bias?
-*  Was there anything you expected to be biased, but the experiments showed fairness in the model outputs?
-*  On the contrary, was there anything you expected to be fair, but the model showed bias?
-
+We will be working with the [[https://colab.research.google.com/drive/1mFZDm28NbKy5oe-ltK6nnnFpIsSoT9U7?usp=sharing|Google Colab Notebook]]. The task consists of experiments with prompting a assistant model for solving the sentiment analysis task and math problem tasks.