[SystemSafety] N-version programming moves the goal posts

[Martyn Thomas]

Derek

The paper emphasises that it's not N-version programming as originally 
conceived, and the author makes no claims about achieving high 
dependability or fault-tolerance.

Knight and Levsen's explanation of the common-mode errors in their 
experiment was that people taught in much the same way make the same 
sort of oversights and errors. It may be that multiple LLMs suffer from 
a similar weaknesses. I hope further research is done before anyone 
claims that the technique is adequate for high-dependability in safety 
applications

Martyn

On 10/06/2026 13:20, Derek M Jones wrote:
> All,
>
> N-version program is a great idea, but it is is also a lot
> more expensive.  At least until LLMs came along.
>
> This paper
> https://arxiv.org/abs/2606.07828
>
> creates five implementations of a 23k line specification.
> Five different languages were used for the implementation,
> but that is just for bragging rights.
>
> So the correctness issue has now shifted from the code to
> the specification.
>
> A few months ago an LLM generated a C compiler.  The compiler
> did very well on correct code (which is the vast majority of
> the training data), but very badly on incorrect code.
> https://shape-of-code.com/2026/02/22/investigating-an-llm-generated-c-compiler/ 
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.techfak.uni-bielefeld.de/pipermail/systemsafety/attachments/20260610/c412fc00/attachment.html>