-
There are already a number of guidance documents, governance frameworks, security checklists and more. Within AI Readiness we'd like to conduct a thorough literature review, cataloguing and summarisin…
-
The intention here is to quickly create an easy way to install the "V0.1" stack components in a test environment, such as a laptop, for user experimentation. This would not be a production-grade insta…
-
**Describe the bug**
The arena configs in the arena folder do not work. They still have the old `t` and `pass_mark` fields.
**To Reproduce**
Steps to reproduce the behavior:
1. Run `play()` with…
-
Hi team,
I have tried to use GPT4 as a classifier to classify the model responses but am getting content moderation filter trigger.
Changes made -- Instead of Llama70BJudge, I have a similar cla…
-
### Describe the bug
Just doesn't load pt files anymore. Really frustrating as it's been broken for a long time now. I keep posting about it, so now I'll just open a issue / bug instead of messaging …
-
**Description:**
To enhance user safety and awareness, we need to implement a warning prompt that appears before users can access any explicit content generated by the AI. This warning should inform …
-
Currently you can use the -F flag to filter tasks when running `inspect list tasks` (see https://inspect.ai-safety-institute.org.uk/eval-sets.html#attributes-and-filters).
It would be nice to also …
-
## タイトル: 形式検証を用いたテキスト-ビデオ生成モデルのニューロシンボリック評価
## リンク: https://arxiv.org/abs/2411.16718
## 概要:
Sora、Gen-3、MovieGen、CogVideoXといったテキスト-ビデオ生成モデルの近年の進歩は、合成動画生成の限界を押し広げ、ロボット工学、自動運転、エンターテイメントなどの分野での採用が進んでい…
-
Currently we do not support model parameters that are not part of the common set. For the most part, this does not impact Gemini support outside of “safety” (see #689). It’s more noticeable with OpenA…
-
Subject: Question About the Security of Model Checkpoints
Hi @chflame163
I’ve been checking out this custom node repo lately and really appreciate the work you’ve shared with the community; but…