# forck.live — 3–9 October 2022 > AI lab announcements, news reports and podcasts. forck.live follows what companies announce about their own products. Press may corroborate an official story; it never leads and never appears as an item of its own. A missing date stays missing. A title is the publisher's words — where a company does not publish in English, an English rendering is listed separately as `Title in English` and named as this site's translation, never as theirs. This file is one issue of the weekly archive: every confirmed first-party announcement of ISO week 2022-W40, 3–9 October 2022, newest first. The window never moves, so this URL always names the same seven days — but a post published inside them and detected later is added when it arrives, so a closed week can still gain entries. Every entry is one first-party announcement, and carries the same eight fields in this order before any others: Company · Announced · Category · Coverage · Announcement · Group · Source · Record `Source` is the company's own page the announcement was traced to and `Record` is this site's page for it. Both are safe to cite. `Coverage` is a number of outlets, or `not counted` — which means this record reached no count of its own, never that nobody wrote about it. `Category` is this tracker's own analysis, and is authoritative wherever it is stated. `not stated` means either that nothing has analysed the post yet or that the analysis placed it outside this taxonomy; neither is a verdict about the announcement. `Announcement` is whether the post is news of something the company did, rather than a how-to, a customer story, a changelog listing or a staffing note. `Group` is which of this record's groups the entry falls in, carried on the entry itself so it holds where the groups are headings and where they are not. One of: `models`, `covered`, `announcements`, `routine`. Anything after `Record` is held detail that only some announcements carry. - This week's page: https://forck.live/week/2022-W40 - The current issue, and the catalogue of the other weeks: https://forck.live/briefing.md - Catalogue: https://forck.live/llms.txt ## Announcements ### Hugging Face — Introducing DOI: the Digital Object Identifier to Datasets and Models - Company: Hugging Face (huggingface.co) - Announced: 2022-10-07 - Category: capability-change - Coverage: not counted - Announcement: yes - Group: announcements - Source: https://huggingface.co/blog/introducing-doi - Record: https://forck.live/items/2146-introducing-doi-the-digital-object-identifier-to-datasets-and-models - Subject: Platform Hugging Face announced a new feature allowing users to generate Digital Object Identifiers (DOIs) for their models and datasets directly from the Hub, in partnership with DataCite. ### LG AI Research — [ICML 2022] Part 3: Neural Combinatorial Optimization studies - Company: LG AI Research (lgresearch.ai) - Announced: 2022-10-05 - Category: research-paper - Coverage: not counted - Announcement: no - Group: routine - Source: https://www.lgresearch.ai/blog/view?seq=258 - Record: https://forck.live/items/4928-icml-2022-part-3-neural-combinatorial-optimization-studies - Subject: EXAONE Three LG AI Research researchers present selected studies from ICML 2022. Part 3, by Hanseul Jeong, focuses on Neural Combinatorial Optimization (NCO) and introduces two studies: LeNSE, a method for large-scale combinatorial optimization using subgraph embeddings and reinforcement learning to navigate subgraphs efficiently. ### Hugging Face — Japanese Stable Diffusion - Company: Hugging Face (huggingface.co) - Announced: 2022-10-05 - Category: new-model - Coverage: not counted - Announcement: yes - Group: models - Source: https://huggingface.co/blog/japanese-stable-diffusion - Record: https://forck.live/items/2147-japanese-stable-diffusion - Subject: Platform rinna Co., Ltd. has developed a Japanese-specific text-to-image model named 'Japanese Stable Diffusion' by fine-tuning Stable Diffusion on Japanese-captioned images. ### Replit — Replit's New Logomark - Company: Replit (replit.com) - Announced: 2022-10-04 - Category: not stated - Coverage: not counted - Announcement: no - Group: routine - Source: https://replit.com/blog/new-logo - Record: https://forck.live/items/18193-replit-s-new-logomark - Subject: Replit Agent Replit has changed its logo to a simplified symbol called the "prompt," which consists of three yellow dots. The company states the new logo is easier to draw, scales well, and aligns with its goal of providing a starting point for creation. The logotype has also been updated to a monospace font. ### LG AI Research — [ICML 2022] Part 2: Self-supervised learning - Company: LG AI Research (lgresearch.ai) - Announced: 2022-10-04 - Category: research-paper - Coverage: not counted - Announcement: no - Group: routine - Source: https://www.lgresearch.ai/blog/view?seq=256 - Record: https://forck.live/items/4930-icml-2022-part-2-self-supervised-learning - Subject: EXAONE This blog post is a research summary written by an LG AI Research researcher, covering selected self-supervised learning papers presented at ICML 2022, with a detailed explanation of the Adversarial Masking for Self-Supervised Learning (ADIOS) method. ### LG AI Research — [ICML 2022] Part 1: Long-Tail Distribution Learning - Company: LG AI Research (lgresearch.ai) - Announced: 2022-10-04 - Category: research-paper - Coverage: not counted - Announcement: no - Group: routine - Source: https://www.lgresearch.ai/blog/view?seq=257 - Record: https://forck.live/items/4929-icml-2022-part-1-long-tail-distribution-learning - Subject: EXAONE LG AI Research published a blog post summarizing long-tail distribution learning research and introducing their PASCL paper presented at ICML 2022. ### Hugging Face — Very Large Language Models and How to Evaluate Them - Company: Hugging Face (huggingface.co) - Announced: 2022-10-03 - Category: developer-tool-release - Coverage: not counted - Announcement: no - Group: routine - Source: https://huggingface.co/blog/zero-shot-eval-on-the-hub - Record: https://forck.live/items/2148-very-large-language-models-and-how-to-evaluate-them - Subject: Platform Hugging Face announces that zero-shot evaluation is now available for any causal language model on the Hub via the Evaluation on the Hub tool, powered by upgraded AutoTrain infrastructure that allows free evaluation of large models.