Tag: AI for Education

Masses of digital text. AIs with a large context window can process much more of it!

Gemini’s Long Context Window – a True Spec Cruncher

Leave a comment

Maybe you’ve noticed that Google’s Gemini has been making gains on ChatGPT lately. Of all its recent impressive improvements, one of the lesser-sung features – at least in AI for Ed circles – is its much enhanced context window.

The context window is essentially how much text the AI can ‘remember’, and work with.  Google’s next model boasts one million characters of this memory, leaving other models – which count their own in the hundreds of thousands – in the dust. It blows open the possibilities for a particular kind of AI task: working with long texts.

Language learners make use of all kinds of texts, of course. But one particularly unwieldy (although hugely useful) type where this new feature could help is the exam spec.

Exam Spec Crunching with AI

Language exam specs are roadmaps to qualifications, listing the knowledge and skills students need to demonstrate linguistic competency. But they have a lot of fine detail that can bog us down.

As a content creator, one thing that challenges me is teasing out this detail into some kind of meaningful arrangement for student activities. There is a mass of vocab data in there. And as systematic as it is, abstract lists of connectives, temporal adverbs and helper verbs don’t make for very student-friendly lesson material.

With a massive text cruncher like Gemini, they are a lot easier to process. Just drag in your spec PDF (I’ve been playing around with the new AQA GCSE German doc), and tease out the material in a more useful format for planning:

Take this German exam spec, and create an outline plan of three terms of twelve lessons that will cover all of the thematic material.

Additionally, it can help in creating resources that cover all bases:

Create a short reading text to introduce students to the exam topic “Celebrity Culture”. It should be appropriate for students aiming for the top tier mark in the spec. In the text, make sure to include all of the prepositions from the prescribed word list.

With a long textual memory, it’s even possible to interrogate the spec after you’ve uploaded it. That’s literally just asking questions of the document itself – and, with that bigger window, getting answers that don’t overlook half the content:

If students have one year to learn ALL of the prescribed vocabulary in the spec, how many words should they be learning a week? Organise them into weekly lists that follow a broadly thematic pattern.

Supersized Context Window – Playing Soon at an AI Near You!

For sure, you can use these techniques on existing platforms straight away. However, due to the smaller context window, results might not always be 100% reliable (although it’s always fun trying!). For the new Google magic, we’ll have to wait just a little longer. 

But from the initial signs, it definitely looks worth the wait!

Gemini’s new supersized context window is available only in a limited released currently, and only via its AI Studio playground. Expect to see it coming to Gemini Advanced very soon!

An illustration of a robot scribe writing AI prompts for ChatGPT or Gemini

Power to the Prompts : Fave Tweaks for AI Worksheets

Leave a comment

Content creation is what AI excels at. That’s a gift to language learners and teachers, as it’s the easiest thing in the world to create a set of prompts to run off original, immersive worksheets.

AI isn’t great at everything, though, which is why our prompts need to tweak for its weaknesses – the things that are obvious gotchas to language folk, but need explaining to a general assistant like AI. Fortunately, in most cases, all you need is an extra line or two to put it right.

Here are five of my favourite prompt-enhancers for worksheets, covering everything from vocab to copyright!

Five Tweaks for Perfect Prompts

Cut Out the Cognates

Cognates are generally great for language learners. As words that are instantly recognisable, they’re extra vocab for free. But for that very reason, they’re not the ones that worksheets should be making a song and dance about. It’s not particularly useful, for example, to have “der Manager” picked out in your German glossary as a key word. Yes, I worked that one out!

Try this in your vocab list prompts, bearing in mind that not all platforms will work equally well with it (Gemini aced it – ChatGPT sometimes gets it):

Don’t list items that are obvious English loanwords or cognates with English.

Highlight the Good Stuff

You know what I mean by the good stuff – those structures and snippets of languages that are really frequent, and really reusable. I like to call them sentence frames – you can learn them, and switch in other words to add to your own linguistic repertoire.

You can ask AI to draw attention to any really pertinent ones in your target language texts:

Highlight (in bold and italics) the most pertinent key words and phrases for the topic, and provide a brief glossary of them at the end.

Make It Colloquial

Vanilla AI can sound bookish and formal. That’s no good as a model for everyday speech, so polish your prompts with a wee push to the colloquial:

Make the language colloquial and idiomatic, in the style of a native speaker.

Include an Answer Key

It might seem obvious, but if you’re making materials for self-study, then you will find an answer sheet indispensable. It’s an often overlooked finishing touch that makes a worksheet truly self-contained:

Include an answer key for all questions at the end.

Covering Your Back With Copyright

Copyright issues have been bubbling in the background for LLMs for some time now. They produce texts based on vast banks on training data, which isn’t original material, of course. But in theory, the texts that pop out of it should be completely original.

It can’t hurt to make that explicit, though. I like to add the following line to prompts, especially if I’m intending to share the material beyond personal use:

Ensure that the text is completely original and not lifted directly from any other source.

 

What are your favourite tweaks to make perfect prompts? Let us know in the comments!

A deck of neon flashcards. Anki cards might not be quite as fancy!

From ChatGPT to Anki : Instant Potted Vocab Decks!

1 Comment

With cutting edge AI galvanising the language learning world, traditional tools like Anki – which would have been considered the leading edge not that long ago – seem well in the shade. But it’s not a question of either-or. Traditional and new tech can work in happy symbiosis to support language learning.

Preparing for a recent high-stakes language mission (OK, island-hopping hol!) to Greece, I wanted to turboboost my Greek vocab. Anki was my tool of choice, of course, but one question remained: where to source new flashcard decks? Large Language Models like ChatGPT and Bing were easy choices for generating topical vocab lists, but how much copy-pasting would that involve? I wasn’t keen on spending hours formatting cards manually.

Thankfully, ChatGPT Plus’ Advanced Data Analysis mode can provide a bridge between old and new. Forget that slightly intimidating title – the main boon is simply that this mode can output a text file. And, given the right format, Anki can take such a text file as an import source. With a bit of prompting prowess, we can automate the whole process – from topic to cards, in one fell swoop. Before long, I had a fresh daily drip-drip of new words and phrases, a real shot in the arm for my Greek pre-trip.

Here’s how to task ChatGPT with the whole job of Anki deck creation. If you don’t have the Plus version, no problem – scroll down for a modified version that works with completely free plans and services!

Automatic Anki Decks – Plus Style

First of all, start a new chat in ChatGPT, and make sure Advanced Data Analysis is selected in the drop-down menu under ChatGPT-4.

Selecting Advanced Data Analysis mode in ChatGPT-4.

Selecting Advanced Data Analysis mode in ChatGPT-4.

Now, we’re ready for our prompt. Like our AI speaking prep worksheets, the beauty of this is just how specific you can make your flashcards. The topic can be as broad or narrow as you like. Here’s a sample prompt to create a French deck on the talking point ‘social issues’:

Hello! I’m learning French, and I’d like you to create an Anki flashcard deck to help me. To import a deck, Anki requires a CSV file format with just a “Front”, “Back” and “Tags” field corresponding to the English. the target language phrase and the part of speech. There is no need for header fields, so the first line should represent the first vocabulary item.
Can you create such an Anki-ready list of 50 flashcard items on the topic “Social Issues” for me, then save it for me as a .txt file I can import into the app?
– Provide a good mixture of essential and useful nouns, verbs, adjectives, and useful phrases / sentence frames (ie., so it’s not just a list of nouns!).
– Provide each term in its dictionary form if appropriate, indicating gender, plural and essential or irregular parts briefly as per convention where applicable.
– Ensure that all terms relate to the contemporary culture of the target language country as much as possible.
– Please draw on resources in the original target language when researching which words will be most useful, cross-referencing with all available data and checking constantly to make sure that the target language for the flashcards is accurate and colloquial, never bookish or unnatural.

Limitations (For Now)

One limitation with the Advanced Data Analysis mode is that it can’t run concurrently with ChatGPT’s now restored web-connected mode, or Browse with Bing. All that means is that it will be relying on its banks of training data for the vocab collation, rather than the web. But in most cases, it shouldn’t make too much difference given the vastness of that data (although it will notify you apologetically about it – see below). We’re waiting for the day – hopefully soon – that OpenAI allows users to run several premium features together.

ChatGPT Plus whirring away creating an Anki deck.

ChatGPT Plus whirring away at an Anki deck. Quirky repartee not as standard, but provided by special request thanks to custom instructions! I like my AI cheeky.

Into Anki We Go

One you have your ChatGPT-infused vocab file ready, you can import it straight into Anki. In the Anki desktop app, head to File > Import, and select the file you saved. The import settings window will pop up, including, crucially, which field matches to which column of your data under Field Mapping. The app guesses correctly for the most part, but occasionally you may need to specify that the third column (part of speech) maps to the tags field.

Importing CSV data into Anki decks.

Importing CSV data into Anki decks.

And that’s it. You should get a brief report of the number of items added, and they’re ready to play with straight away. Instant, fresh vocab decks in seconds!

No ChatGPT Plus? No problem!

Now, the above is all very well if you have ChatGPT Plus. Many platforms lack the file output side of things. But you can still get them do the heavy work of vocab-hunting and file-formatting; all you need to do is the final copy-paste-save.

Here’s how to alter the prompt for plain old vanilla ChatGPT and Bing, coaxing it to provide Anki-ready output. I’ve also made the format a little clearer, which might help if you’re using slightly older models like ChatGPT-3.5.

Hello! I’m learning French, and I’d like you to create an Anki flashcard deck to help me. To import a deck, Anki requires a CSV file format with just a “Front”, “Back” and “Tags” field corresponding to the English. the target language phrase and the part of speech.
Can you create such an Anki-ready list of 25 flashcard items on the topic “Driving a Car” for me? Output the CSV data as formatted as code so I can easily copy-paste into a text file for Anki.
– Don’t include header fields in the CSV – the first line of your output should be the first vocabulary item (ie., car,la voiture,noun).
– Provide a good mixture of essential and useful nouns, verbs, adjectives, and useful phrases / sentence frames (ie., so it’s not just a list of nouns!).
– Provide each term in its dictionary form if appropriate, indicating gender, plural and essential or irregular parts briefly as per convention where applicable.
– Ensure that all terms relate to the contemporary culture of the target language country as much as possible.
– Please draw on resources in the original target language when researching which words will be most useful, cross-referencing with all available data and checking constantly to make sure that the target language for the flashcards is accurate and colloquial, never bookish or unnatural.

Your platform should spool out some easily copiable code. Simply paste this into a text file, save, and import into Anki.

Even using 3.5, I got some great results featuring practical, useful vocabulary sets.

Creating Anki decks with the free ChatGPT3.5 model.

Creating Anki decks with the free ChatGPT3.5 model.

Experiment, Experiment, Experiment!

As with all AI prompts, it’s worth experimenting with everything to tweak, improve and get the absolute best out of it. The number of cards, the mix of words and phrases, the source of the material – make it your own. When you have it just right, you can create cards for your own, or your students’ learning, in seconds.

Oh, and don’t forget to save your perfect prompts somewhere you can copy-paste them from later, too!

If you’re keen for more artificial intelligence tips to boost your learning, please check out my book AI for Language Learners. It’s packed with practical examples to fuel your linguistic adventures!