search

ls1intum / Athena

A system to support (semi-)automated assessment of exercises, built on ls1intum/Athena
https://ls1intum.github.io/Athena/
6 stars 1 forks source link

Multiple Selectable Approaches for Text (Chain of Thought with Prompt Chaining) #350

Closed EneaGore closed 5 days ago

EneaGore commented 1 month ago

Motivation and Context

This PR provides a new structure for different approaches within the text llm module. It already adds a new approach "chain of thought".

Description

The standard approach has been moved in a folder named basic_approach. Each approach has its own folder so they can be easily added, or deleted in case it is not needed anymore. The approach controller and config need to be adjusted when a new approach is added or removed.

Steps for Testing

  1. Select text llm module
  2. Observe the new approach selector, pick either one.
  3. Confirm that the request is processed and that the correct one was called (either use langsmith locally or see that, for example, in the case of chain of thought with chain prompting there are 2 llm calls in the meta data output)

    Testserver States

    [!NOTE] These badges show the state of the test servers. Green = Currently available, Red = Currently locked Click on the badges to get to the test servers.

Screenshots