Mnemosphere

Scite is an AI research assistant that provides answers backed by scientific citations. It draws from a large database of full-text articles, including paywalled sources. The tool helps researchers explore topics, build reference lists, and verify claims. Scite also offers customizable settings and filters to refine research output.

Qeios is an open science platform that provides free, rapid, and transparent open access publishing. It offers unbiased open peer review for researchers. Qeios connects AI labs and startups with specialized experts for evaluations and collaborations. Qeios Quorum assists with human evaluations, RLHF, and high-impact collaborations.

REP is an ipython-based environment designed for conducting data-driven research. It emphasizes consistency and reproducibility in machine learning experiments. REP provides a toolbox for data scientists and machine learning engineers to streamline their research workflows. It is an open-source tool.

Quantitative-Research-Projects is a collection of quantitative finance research projects. It covers sector rotation, multi-factor models, and AI-driven strategies. The projects span high, mid, and low frequencies. Each project includes full code and analysis and is continuously updated.

RTDL (Research on Tabular Deep Learning) is a collection of papers and packages on deep learning for tabular data. It provides resources for researchers and practitioners. Users can follow announcements on new projects. The tool supports exploration of deep learning models for tabular data.

rome is a tool for locating and editing factual associations in GPT models. It implements Rank-One Model Editing (ROME) on auto-regressive transformers. The tool currently supports OpenAI's GPT-2 XL (1.5B) and EleutherAI's GPT-J (6B). It is designed for GPU-only operation.