Tag
Lophius is a new hybrid code/GUI research system for language models that runs inside a notebook, aiming to reduce boilerplate and streamline tasks like model inspection, tokenizer analysis, and inference.
NotebookLlaMa is a fully open-source alternative to NotebookLM, built on LlamaCloud with an MCP server and Streamlit interface for turning source material into a research workspace.
Suzgun and colleagues announce string2string Studio, an open-source in-browser platform for alignment, comparison, search, and evaluation of strings and sequences.
The author built A2Anet, a free protocol/tool based on Google's A2A that lets two or more agents collaborate via a shared link with private sandboxes, and asks the community for feedback.
A compiler that converts computation graphs defined in Python into the weights of a standard transformer architecture (Phi-3) without any training, enabling vanilla HuggingFace deployment with no custom code.
Awesome-LLM-Survey is a topic-organized GitHub repository collecting LLM survey papers and project links to help researchers quickly find overviews across training, prompting, modalities, and applications.
Google renames NotebookLM to Gemini Notebook, adding a secure cloud computer for native code execution and deeper integration with the Gemini app and Google Search.
MafiaScope is an open testbed that uses the social deduction game Mafia to probe LLM agents' beliefs non-invasively and in real-time, enabling fine-grained analysis of machine Theory of Mind through structured probe questions, interactive visualization, and counterfactual replay.
Cito is a hybrid academic search tool over 236 million papers, designed for AI agents.
OpenScience is an open-source alternative to Claude Science, supporting multiple AI models and over 250 research skills, with native Atlas integration for reproducible research graphs and user-controlled infrastructure.
Claude Science is a tool for early-stage research that automatically retrieves, deduplicates, clusters literature and generates research maps and trend reports, which is very helpful for exploring new fields.
Introducing the PDF2zh translation plugin for Zotero, which supports one-click translation of English PDF papers into Chinese, preserving formulas and layout, providing side-by-side Chinese-English comparison, and supporting multiple translation engines.
The author shares their positive experience with NotebookLM and teases five prompts for efficiently processing large amounts of text.
DoTadda is an AI-powered research operating system for investors, combining knowledge management with AI to help analysts and portfolio managers store, search, and chat with their research documents and earnings call transcripts.
AI-research-feedback is an academic paper review skill for Claude Code. It checks grammar, coherence, formulas, figures, and argument flaws through six parallel agents, supports specifying journals to simulate reviewers, and finally generates a structured review report.
Manus AI now integrates with Mobbin, allowing users to automate design research across top products.
A developer introduces Primnox, a desktop AI tool that runs a local NER model to scrub personally identifiable information before sending data to the cloud, featuring a self-building knowledge graph, deep research mode, and markdown notes with AI actions.
Stanford's STORM research method uses multiple AI agents with different perspectives to produce more organized reports. The author implemented it as a free Claude skill that generates verified HTML briefings.
Paper Lifecycle is a suite of Codex skills designed for academic writing, comprising two skills: Review Revision and Rebuttal Response, which help researchers improve paper quality and handle reviewer feedback.
Medical students are reportedly using a popular research tool to produce misleading studies, raising concerns about academic integrity and research quality.