toxicity
hate-speech-and-offensive-language
toxicity | hate-speech-and-offensive-language | |
---|---|---|
11 | 2 | |
166 | 750 | |
0.0% | - | |
0.0 | 1.9 | |
almost 2 years ago | 11 months ago | |
Jupyter Notebook | ||
MIT License | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
toxicity
-
Perhaps It Is a Bad Thing That the Leading AI Companies Cannot Control Their AIs
I'm a PM at a human data company (https://www.surgehq.ai) that helps the large language model companies ensure their models are safe (we're the “clever prompt engineers” who helped Redwood assess their model performance).
We actually just published a blog today that includes our perspective on building “AI red teams” and best practices for AI alignment/safety: https://www.surgehq.ai/blog/ai-red-teams-for-adversarial-tra...
-
30% of Google's Emotions Dataset Is Mislabeled
I'd love to chat. Want to reach out to the email in my profile? I'm the founder of a much higher-quality data startup (https://www.surgehq.ai), and previously built the human computation platforms at a couple FAANGs.
We work with a lot of the top AI/NLP companies and research labs, and do both the "typical" data labeling work (sentiment analysis, text categorization, etc), but also a lot more advanced stuff (e.g., training coding assistants, evaluating the new wave of large language models, adversarial labeling, etc -- so not just distinguishing cats and dogs, but rather making full use of the power of the human mind!).
-
Building a No-Code Toxicity Classifier – By Talking to GitHub Copilot
> Rather than operating under a strict definition of toxicity, we asked our team to identify comments that they personally found toxic.
[0]: https://github.com/surge-ai/toxicity
-
Ask HN: Who is hiring? (January 2022)
Love language? So do we, and our mission is to infuse AI with that same love. At Surge, we're building the human infrastructure to power NLP — from detecting hate speech, to parsing complex documents, to injecting human values into the next wave of language models. Our first product is a platform that helps ML teams create amazing, human-powered datasets to train AI in the richness of language. We're a team of former Google, Facebook, and Airbnb engineering leads, and we work with top companies at the forefront of machine learning. Our tech stack is Ruby on Rails, React, and Python. We’re rapidly growing, and we're looking for full-stack engineers to join the team and develop our product. To apply, please email [email protected] with a resume and 2-3 sentences describing your interest in Surge. We love personal projects and writings too!
More information: https://www.surgehq.ai/about#careers
A blog post explaining the problems we are working to solve: https://www.surgehq.ai/blog/the-ai-bottleneck-high-quality-h...
- The Toxicity Dataset – building the largest free dataset of online toxicity
- [Free] The Toxicity Dataset — building the world's largest free dataset of online toxicity [Github]
- The Toxicity Dataset — building the world's largest free dataset of online toxicity
- The Toxicity Dataset (1000 social media comments) — any ideas for interesting visualizations? [github]
- The Toxicity Dataset - free dataset of online toxicity (Github) - could be used for interesting portfolio projects
- The Toxicity Dataset — free dataset of online toxicity (Github)
hate-speech-and-offensive-language
-
How to make a class column for a classifier from sentiment analysis results?
I've used NRCLex to perform sentiment analysis on some Twitter data. I have hate speech classifier code (https://github.com/t-davidson/hate-speech-and-offensive-language/blob/master/classifier/final_classifier.ipynb) I want to pass the dataset through, but before I can I need to have a "class" column for the model. For those not familiar, NRCLex returns scores for 10 emotions: anticipation, joy, anger, fear, surprise, disgust, positive, negative, sadness and trust. The table looks like this (letters denoting emotions):
-
Where do we go from here and who is going to step up to help us?
Some of this exists, and both Quora and Facebook (among others) use it extensively. Both hate speech and porn are good targets for machine learning. It needs supervision, but it can take a lot of load off human moderators.
Open source implementations exist, e.g.:
https://github.com/t-davidson/hate-speech-and-offensive-lang...
I suspect more message board will want to start applying these sooner rather than later. Most have already figured out that they need anti-spam tools, rather than it coming as a surprise when they roll things out and it fills up with bots. The technology is similar.
You mention being able to share that information across boards, and I don't know of any widespread implementation of that. You can, at least, let somebody else handle your authentication, which slightly slows their ability to create new accounts when you blacklist one. I'd like to see those sites distinguish "aged" accounts, so that it at least takes some effort or cost to use a new account.
What are some alternatives?
seldon-core - An MLOps framework to package, deploy, monitor and manage thousands of production machine learning models
cia - 🐱💻 CIA Factbook data analysis and dataset reconstruction, modification, and tuning go here.
zotero - Zotero is a free, easy-to-use tool to help you collect, organize, annotate, cite, and share your research sources.
Tegridy-MIDI-Dataset - Tegridy MIDI Dataset for precise and effective Music AI models creation.
Fleet - Open-source platform for IT, security, and infrastructure teams. (Linux, macOS, Chrome, Windows, cloud, data center)
ThoughtSource - A central, open resource for data and tools related to chain-of-thought reasoning in large language models. Developed @ Samwald research group: https://samwald.info/
zenml - ZenML 🙏: Build portable, production-ready MLOps pipelines. https://zenml.io.
airline-sentiment-streaming - Streaming with Airline Sentiment. Utilizing Cloudera Machine Learning, Apache NiFi, Apache Hue, Apache Impala, Apache Kudu
datapane - Build and share data reports in 100% Python
100daysofpractice-dataset - Data from Instagram posts with the hashtag #100daysofpractice.
deno - A modern runtime for JavaScript and TypeScript.
hashformers - Hashformers is a framework for hashtag segmentation with Transformers and Large Language Models (LLMs).