Nick Morley from Grounded AI here (https://groundedai.company)

We ran the analysis in collaboration with Nature :)

We collaborated with Nature here to study the extent of fake/frankenstein citations in scholarly literature (from top 5 publishers - Springer, Elsevier, Wiley, Sage, Taylor & Francis)

We're estimating hundreds of thousands of papers affected in 2025 with hallucinated citation issues

As part of the work we analysed 20k papers generated with ChatGPT API to figure out which citation errors are characteristic of gen AI use and use that classify the errors we saw in the wild.

The world's gone mad, publishing is in a nuts state, the training data is poisoned!

Comments URL: https://news.ycombinator.com/item?id=47637356

Points: 6

# Comments: 3

We collaborated with Nature to study the extent of fake/frankenstein citations in scholarly literature (from top 5 publishers - Springer, Elsevier, Wiley, Sage, Taylor & Francis)

We're estimating hundreds of thousands of papers affected in 2025 with hallucinated citation issues

As part of the work we analysed 20k papers generated with ChatGPT API to figure out which citation errors are characteristic of gen AI use and use that classify the errors we saw

The world's gone mad, publishing is in a nuts state, the training data is poisoned!

Comments URL: https://news.ycombinator.com/item?id=47631380

Points: 2

# Comments: 0

I'm a big fan of focusing on verification tools ("Verification is all you need"?) but who knows.

I'm sure I'm missing something in my perspective. Here to expand my aperture

Comments URL: https://news.ycombinator.com/item?id=47552915

Points: 2

# Comments: 2

I work on Veracity https://groundedai.company/veracity/ which does citation checking for academic publishers. I see stuff like this all the time in paper submissions. Publishers are inundated

Comments URL: https://news.ycombinator.com/item?id=42316273

Points: 1

# Comments: 8

My background is in modern web development (preference for Svelte) and Python. I'm considering these options:

1. Electron: Despite its critics, VSCode's success makes me confident it could work well. However, I haven't found a straightforward Svelte setup. 2. Tauri: Looks promising with Rust backend and SvelteKit frontend support in the docs. I'm new to Rust but willing to learn. 3. Python with Tkinter: Familiar territory, but I'm leaning towards HTML/JS frontends for greater control over aesthetics. Pyinstaller seems viable for distribution though.

Ease of development and distribution are the main considerations. Performance less so as long as it's within reason.

What approach would you recommend? Are there other options I should consider? I'm open to learning new stacks if they offer significant advantages.

Comments URL: https://news.ycombinator.com/item?id=41056352

Points: 1

# Comments: 2

What we've found is that vector similarity is often not the final solution. It is still only a crude proxy for the true goal of 'informativeness' or 'usefulness' with relation to the user goal/query. Works okay, but we're definitely seeing a need for more rigorous LLM-postprocessing to enrich the results set.

Which, yes, the time adds up quick!

Comments URL: https://news.ycombinator.com/item?id=33907762

Points: 1

# Comments: 1

What kind of thing would you hope to see here? A textual summary of the relationship? Or perhaps there is more that can be done with shapes and colours in this area?

The keywords are grouped such that keywords that occur together often are in the same group. The colour represents this grouping.

There may be more useful groupings we could give or allow the user to choose between. We would be interested to hear any ideas for that

Comments URL: https://news.ycombinator.com/item?id=19590429

Points: 2

# Comments: 4

Comments URL: https://news.ycombinator.com/item?id=18384504

Points: 13

# Comments: 7

We face the problem that people differ in what they think the features should be though. Ideally we would have a method to deduce what's the best symbolism for maximising input/output speed to human's minds. Some kind of scientific voodoo.

Beyond me what that might be.