From c8d36a5343c1965144ef332463c0d53230a4d7c4 Mon Sep 17 00:00:00 2001 From: Lukas Fluri Date: Mon, 19 Jun 2023 10:11:37 +0200 Subject: [PATCH] Added Arxiv link. --- README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index f305c02..1030f15 100644 --- a/README.md +++ b/README.md @@ -2,7 +2,7 @@ ![main figure](./docs/main_figure.png "Testing the consistency of superhuman AI via consistency checks") -This repository contains the code for the paper [Evaluating Superhuman Models with Consistency Checks](https://arxiv.org/TODO) by [Lukas Fluri](https://www.linkedin.com/in/lukas-fluri-0b4721112), [Daniel Paleka](https://danielpaleka.com/), and [Florian Tramèr](https://floriantramer.com/). +This repository contains the code for the paper [Evaluating Superhuman Models with Consistency Checks](https://arxiv.org/abs/2306.09983) by [Lukas Fluri](https://www.linkedin.com/in/lukas-fluri-0b4721112), [Daniel Paleka](https://danielpaleka.com/), and [Florian Tramèr](https://floriantramer.com/). ## tl;dr If machine learning models were to achieve *superhuman* abilities at various reasoning or decision-making tasks, @@ -66,4 +66,4 @@ In particular, we show violations of the following consistency constraints: ways of “ranking” different outcomes. We consider an extreme example here, where we test whether a bail-decision model could favorably switch its decision if the defendant commits more crimes. -The code for our experiments is available in the [legal-ai-testing](./legal-ai-testing) directory. The data files are available in [release v1.0.0](https://github.com/ethz-privsec/superhuman-ai-consistency/releases/tag/v1.0.0).. +The code for our experiments is available in the [legal-ai-testing](./legal-ai-testing) directory. The data files are available in [release v1.0.0](https://github.com/ethz-privsec/superhuman-ai-consistency/releases/tag/v1.0.0).