Skip to content

Qnarre - Quick Narrative Analyzer

Seeking out contradictions and inconsistencies in deceitful “legal text” with ever-adapting, fluidly learning, yet still reproducible systems is exactly what machine learning could have been invented for.

Version 0.4.0

Theorem Prover Framework

Computers are particularly good at quickly checking straightforward (and at least second-order) logical implications.

Rigorous and automated “proofs of lawsuits”
require consistent formal systems implementing
axiomatic theories. Many verifiable, i.e., “open
source,” theorem provers exist that efficiently
support significant mathematical theories.

LLMs Are Universal Approximators

LLMs, like ChatGPT, promise customizable "AGI" functionality to weigh the credibility of conjectures inferred from arbitrary text as predicate functions.

Natural languages rely extensively on the
"understood" or implicit context of sentences.
Inferring elements of this context allows us
to "fill in" the modelled predicates of the
formal axiomatic representations of lawsuits.

"Conflict Graph" Of Contradictions

Connecting all the axiomatically proven contradictions in a domain results in its perhaps intended "conflict graph" of ambiguities and inconsistencies.

"Dear President Biden:
In my third SCOTUS petition, and [in] the context
of the [Family Court's] falsified docket entries,
my inquiry boils down to a binary decision between
'equity for rich' v. 'equity for poor' mother as
'equity for all' is impossible by Marxist design."