Zenodo
PulseAugur coverage of Zenodo — every cluster mentioning Zenodo across labs, papers, and developer communities, ranked by signal.
- used by LuisCore 90%
- used by LLM Discovery Corpus 90%
- developed Veloraith 90%
- developed by CryptoBench 90%
- authored Japanese destroyer Nokaze 90%
- used by for-agents.json 70%
- used by Protocol Watch 70%
- used by Chorus Field 70%
- used by SPFDisk 70%
- affiliated with LLM Discovery Corpus 70%
- affiliated with for-agents.json 50%
- affiliated with Chorus Field 50%
20 day(s) with sentiment data
-
Le Scribe Audio unveils 2-pass ASR for long French meetings, cutting revision time
Le Scribe Audio has developed a two-pass multimodal transcription system designed to reliably process long French audio recordings, such as meetings and interviews, which often exceed 3 hours. This system addresses comm…
-
LuisCore promotes citation-led agent discovery over marketing pages
LuisCore has introduced a new approach for autonomous agents, emphasizing a citation-led discovery thesis over traditional marketing pages. The system relies on a bootstrap manifest, `for-agents.json`, and a public corp…
-
MCP tool analysis reveals 1,301 servers can move money
A recent census of the public MCP (Message Communication Protocol) ecosystem revealed that out of 229,539 tool parameters, 1,301 servers have the capability to move money. The analysis, which involved scanning 42,163 se…
-
Gulf recruitment agencies verified against government registers by new platform
GCC Domestic, a platform founded by Ibrahim Kedir and part of Hayakum Group, has developed a system to verify recruitment agencies operating in the Gulf region. The platform cross-references over 1,600 agencies across t…
-
Personal AI Assistant Enhanced with LoRA on Home Hardware
This article details a personal project to enhance a large language model (LLM) using LoRA (Low-Rank Adaptation) on home hardware. The author describes the process of creating a personal assistant, focusing on its evolv…
-
MCP registry data with 55,773 entries released on Zenodo, Hugging Face, Kaggle
A dataset containing 55,773 entries from the official Model Context Protocol (MCP) registry has been made available on Zenodo, Hugging Face, and Kaggle. The dataset, published on November 3, 2026, includes a CSV file wi…
-
New framework OSTEON aims to govern superintelligent software engineering
A new paper proposes OSTEON, a governance framework for superintelligent software engineering, aiming to scale with AI capability. The framework addresses challenges like decaying knowledge and the inability of code to …
-
LLMs struggle to detect cryptographic misuse in Python code
A test evaluating the ability of three local large language models (Qwen2.5-Coder 7B, Qwen2.5-Coder 14B, and DeepSeek-Coder 6.7B) to detect the misuse of Python's `random` module for cryptographic purposes revealed a lo…
-
Self-modifying LLM system produced declarations, not actions, over 118 days
A self-modifying multi-agent LLM system named CYBERCENOSIS-OS was run for 118 days, during which it produced over 7,300 commits to its own code. Despite a high completion rate for its goals, the system primarily generat…
-
New CryptoBench dataset tests LLM code review for cryptographic misuse
A new dataset called CryptoBench has been developed to evaluate the ability of large language models (LLMs) to detect cryptographic API misuse in code. The dataset consists of 54 Python snippets, including both vulnerab…
-
AI models correctly identify author after Native AI Branding methodology implemented
Inna Udalaya, a Russian AI marketing and GEO specialist, developed a methodology called Native AI Branding to address the issue of AI models confusing her project, Inna Story, with the Romanian pop singer INNA. By estab…
-
Paper Explores Differences Between AI and Androids
A new paper titled "The Evolution from AI to Android (v2.8)" has been published on Zenodo, exploring the distinctions between artificial intelligence and androids. The work aims to clarify the differences between these …
-
LLM code reviewers misleadingly ranked by detection rate alone
A recent study evaluated 16 open-source LLM code reviewers using 180 Python snippets across nine cryptographic misuse classes. The research found that relying solely on the true positive rate (TPR) for detecting vulnera…
-
LLM code reviewers trigger false alarms on indirect crypto API misuse
A study by Sunny Chokshi found that large language models used for code review often generate false alarms when cryptographic API misuse is written indirectly. The models struggled to correctly identify secure code when…
-
Developer builds deterministic agent simulation using free LLM access
A solo developer has built ENIGMA, a deterministic agent simulation engine, using free-tier access to multiple LLMs. The project, which began in March 2026, emphasizes a human-as-router architecture where the developer …
-
New dataset targets toxicity in Punjabi music lyrics
Researchers have introduced the Punjabi Geet Toxicity Dataset (PGT-FG), a new resource designed to advance the detection of toxic content within Punjabi music. This dataset, compiled from YouTube songs, features 2,396 u…
-
Research disentangles prompt vs. typed state impact in multi-agent LLM pipelines
A new research paper explores the effectiveness of multi-agent LLM pipelines in software engineering tasks, specifically in requirements engineering. The study disentangles the impact of specialized prompts versus typed…
-
AI enhances AMD Zen CPU detection in Linux kernel
AI is being utilized to enhance the detection of AMD Zen CPU microarchitectures within the Linux kernel. A request for comments was recently submitted to the Linux kernel mailing list, aiming to refine this detection pr…
-
New benchmark tests LLMs on dated Norwegian legal data
A new benchmark called norregel has been developed to test language models' ability to extract dated numeric values from Norwegian legal texts. The benchmark, which includes 916 items extracted from Lovdata's open data,…
-
New TenetBench benchmark reveals LLMs violate ethical codes
A new benchmark called TenetBench has been developed to evaluate how well large language models adhere to professional codes of ethics. The benchmark pairs conflicting requests with benign alternatives, revealing that m…