Scribd
Upload
30 views3 pages

ITNPAI1 Assignment - 1553606962

Uploaded by

h58991888
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
30 views3 pages

ITNPAI1 Assignment - 1553606962

Uploaded by

h58991888
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

ITNPAI1 Assignment: LLM

15th April 2024

Over the last three practical sessions, you will work on developing an LLM-based chatbot. You will be
provided with the basic development environment in a Google Colab Notebook, and will be creating the
chatbot by designing a comprehensive Prompt controlling the bot persona, dialogue strategies, and
specialised information access. You will need intensive testing for the chatbot during the practicals, which
will generate multiple dialogues. More information will be released during the practical sessions.

After these sessions, you will have to submit:

1. The Prompt you have developed (standalone, in its original Notebook cell, without any other code
content). [30% Marks]
a. any other document (.txt or .pdf) dynamically loaded into the prompt (if applicable)
2. A .txt log of the best dialogue produced with the chatbot (with either yourself as a user or another
tester), with the bot and the user utterances clearly identified. [40% Marks]
3. A short report (3,000 words exclusive of figures, footnotes and references, with a 10% tolerance)
describing your application, with a critical appraisal of performance. Should you decide to LLM-
generate your report, you will remain entirely responsible for its contents. Should contents of an
LLM-generated report contradict the results produced this might lead to a lower grade, without any
option to appeal on LLM grounds. [30% Marks]

You won’t be allowed to depart from the development environment (Gpt +/- Langchain). Your Prompt
should be executable in the original Notebook provided to you, even if you later modify some of the
Notebook code. Failure to run your Prompt in the default environment will result in a 33% penalty.

Dialogue length will be limited to 25 dialogue turns (50 utterances). There may be limitations on the size
of text files imported into your prompt, if applicable: these will be communicated during tutorials.

Submission
You have to submit the aforementioned files on Canvas. Please, make sure to submit your assignment
before Monday the 15th of April.

Plagiarism
Work which is submitted for assessment must be your own work. All students should note that the
University has a formal policy on plagiarism which can be found at:

https://www.stir.ac.uk/about/professional-services/student-academic-and-corporate-services/academic-
registry/academic-policy-and-practice/quality-handbook/assessment-and-academic-misconduct/#eight
Plagiarism means presenting the work of others as though it were your own. The University takes a very
serious view of plagiarism, and the penalties can be severe (ranging from a reduced grade in the assessment,
through a fail for the module, to expulsion from the University for more serious or repeated offences).
Specific guidance in relation to Computing Science assignments may be found in the Computing Science
Student Handbook. We check submissions carefully for evidence of plagiarism, and pursue those cases we
find.

Late submission
If you cannot meet the assignment hand-in deadline and have good cause, please ask for an extension on
Canvas (up to 7 days of extension). If you need a longer extension, you may need to apply for extenuating
circumstances. Coursework will be accepted up to seven days after the hand-in deadline (or expiry of any
agreed extension), but the mark will be lowered by three marks per day or part thereof. After seven days
the work will be deemed a non-submission and will receive an X.

Generative AI
For this assignment, the ethical and intentional use of Generative Artificial Intelligence (Gen AI) Tools is
permitted, beyond the regular use of Gen AI within the assignment work itself, which extends, for instance
to the analysis of Dialogues’ properties.

Should you decide to use it also to assist in the preparation of the report, as stated above: you remain entirely
responsible for the final contents, including, but not limited to: accuracy of statements; non-contradicting
statements; readability and style; accuracy and relevance of cited references.

Whenever AI tools are used you should:

• Cite as a source, any AI tool used in completing your assignment. The library referencing
guide should be followed.
• Acknowledge how you have used AI in your work.

Using AI without citation or against assessment guidelines falls within the definition of plagiarism or
cheating, depending on the circumstances, under the current Academic Integrity Policy, and will be treated
accordingly. Making false or misleading statements as to the extent, and how AI was used, is also an
example of “dishonest practice” under the policy. More details below.

Note on Avoiding Academic Misconduct


Work which is submitted for assessment must be your own work. All students should note that the
University has a formal policy on Academic Integrity and Academic Misconduct (including plagiarism)
which can be found here.

Plagiarism: We are aware that assignment solutions by previous students can sometimes be found posted
on GitHub or other public repositories. Do not be tempted to include any such code in your submission.
Using code that is not your own will be treated as “poor academic practice” or “plagiarism” and will be
penalized.

To avoid the risk of your own work being plagiarised by others, do not share copies of your solution, and
keep your work secure both during and after the assignment period.

Collusion: This is an individual assignment: working together with other students is not permitted. If
students submit the same, or very similar work, this will be treated as "collusion" and all students involved
will be penalized.

Contract cheating: Asking or paying someone else to do assignment work for you (contract cheating) is
considered gross academic misconduct, and will result in termination of your studies with no award.

You might also like