Add post on AI review tutorials #331
Conversation
✅ Deploy Preview for tourmaline-marshmallow-241b40 ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
There was a problem hiding this comment.
Thanks for the proposed blog! 🙌
I suspect this won't take much of your time to edit. Main priority is to clearly denote what text is LLM generated, as that is now not clear enough. Also, the blog talks about "we" which makes it seem as if we collectively as community have decided this. I encourage revising this to denote what you as an author are doing/have done.
|
|
||
| A challenge with LLMs trained for general use is finding domain-specific tasks where they can add sufficient value beyond existing human input. Tasks like providing early sense checking and tailored feedback, particularly from differing perspectives, therefore has potential to overcome common bottlenecks in developing training materials (e.g. providing initial comments and flagging obvious issues while waiting for more detailed human feedback). | ||
|
|
||
| As Epiverse-TRACE training materials continue to develop, we plan to explore further uses beyond simple first-pass reviews. For example, LLMs are well suited to synthesising qualitative feedback, increasing the range of insights that can be collected and summarised from learners as we move into beta testing. We also hope to identify opportunities where LLMs can help supplement the learner experience, as demonstrated by emerging tools like [RTutor](http://rtutor.ai/) for descriptive plotting functionality in R, which combines generation of code in response to user queries with translation into shiny outputs. |
There was a problem hiding this comment.
| As Epiverse-TRACE training materials continue to develop, we plan to explore further uses beyond simple first-pass reviews. For example, LLMs are well suited to synthesising qualitative feedback, increasing the range of insights that can be collected and summarised from learners as we move into beta testing. We also hope to identify opportunities where LLMs can help supplement the learner experience, as demonstrated by emerging tools like [RTutor](http://rtutor.ai/) for descriptive plotting functionality in R, which combines generation of code in response to user queries with translation into shiny outputs. | |
| As Epiverse-TRACE training materials continue to develop, I plan to explore further uses beyond simple first-pass reviews. For example, LLMs are well suited to synthesising qualitative feedback, increasing the range of insights that can be collected and summarised from learners as we move into beta testing. We also hope to identify opportunities where LLMs can help supplement the learner experience, as demonstrated by emerging tools like [RTutor](http://rtutor.ai/) for descriptive plotting functionality in R, which combines generation of code in response to user queries with translation into shiny outputs. |
Using the we makes it seem as if this is an Epiverse TRACE decision, which it is not as far as I know.
There was a problem hiding this comment.
We (@avallecam and I) are looking at further LLM use cases (e.g. to support delivery as well as design). Fine to keep as a general 'we' I think, as discussions have been happening across team (and with external collaborators).
There was a problem hiding this comment.
That makes sense. Could you reflect @avallecam as co-author in the YAML? Otherwise it gets confusing.
There was a problem hiding this comment.
Thanks, have now added @avallecam
Co-authored-by: Chris Hartgerink <[email protected]>
Co-authored-by: Chris Hartgerink <[email protected]>
Highlight LLM generated text
|
Thanks for the review - have made some edits that hopefully address. |
|
OK to now merge? |
|
Thanks @adamkucharski 🙏 Happy to merge this. |
This PR adds a blog post describing options for use of LLMs to review open access training materials 'in character'.
RRight before merging:
datefield has been updatedblueprintsto link to this post_freeze/folder is up-to-date