Tag
哈佛物理教授 Matthew Schwartz 借助开源工具 BootLoops 与 Claude，在三个月内跨 18 个学科产出 36 篇研究手稿，其中多篇计算此前从未被完成。该工具由 Anthropic 资助、以 MIT 协议开源，可与 Claude、Gemini 或 ChatGPT 等多种模型配合使用。
Harvard physicist Matthew Schwartz, in an Anthropic Science Blog guest post, argues that current LLMs face an "impedance mismatch" with scientific research and describes BootLoops, a Claude-built toolkit for exact quantitative calculations applied across fields like ecology and population genetics.
The tweet lists 10 open-source GitHub projects that provide memory layers for AI agents to maintain state across sessions, preventing them from starting from zero each time.
DS-Lighting is a unified harness toolkit that makes harness design explicit for data-science automation, improving reproducibility, comparability, and reliability in end-to-end workflows.
reverse-skill is a GitHub tool that provides structured workflows and routing for cybersecurity tasks like APK analysis and CTF challenges, specifically designed for AI agents to enhance efficiency and consistency.
Mistral introduces Agentic Search, a multi-step retrieval loop that enhances AI system accuracy and efficiency in navigating complex documents, reducing latency and token use compared to traditional RAG methods.
Usenet Archive Toolkit provides a set of tools to process and search Usenet messages into a coherent offline archive, offering fast access, deduplication, spam filtering, and UTF-8 transcoding.
A new toolkit enables fine-tuning the tiny Inflect-Nano/Micro TTS models on custom voice and language, supporting warm-start, resumption, and export to PyTorch/ONNX.
WeNet is an open-source end-to-end speech recognition toolkit that unifies training code, pretrained models, and deployment runtimes in one project, supporting multiple model architectures and runtime backends.
Mantis Skills is a portable toolkit from Google for building security review harnesses using coding agents, enabling automated vulnerability detection, reproduction, and patching across various codebases.
AngelSlim/Hy3-GGUF is a toolkit for quantizing and deploying Hy3 large language models on llama.cpp, featuring MTP self-speculative decoding and a thinking/tool-call parser for efficient inference.
FairSelect is a toolkit for systematically evaluating fairness mitigation strategies across preprocessing, inprocessing, and postprocessing stages, supporting intersectional subgroup evaluation and comparison of fairness-utility tradeoffs.
Simulstream is an open-source framework for evaluating and demonstrating streaming speech-to-text translation systems, supporting both incremental and re-translation decoding on long-form speech with fine-grained logging and an interactive web interface.
GitHub released SpecKit, an open-source toolkit that addresses a weakness in AI coding by enforcing a spec-first workflow, turning product specs into executable development contracts for AI agents.
iFixit announces the Megalodon Driver Kit, a $34.95 toolkit designed for household repairs and DIY projects. It includes a new swivel-grip screwdriver and 16 long-shank bits in a compact case.
ManusAI launches a toolkit for small businesses and solopreneurs, including connectors and resources, with live sessions to help users implement it.
Introducing Nub, an all-in-one JavaScript toolkit that augments Node.js instead of replacing it.
Chameleon Ultra is a compact, flashdrive-sized NFC toolkit for RFID research and hacking, with compatible applications and authorized distributors worldwide.
The author built an orientation toolkit that helps AI models orient themselves (e.g., location, context) before reasoning, reducing search waste but increasing verification waste, leading to safer and more efficient performance.
TSDuck is a free and open-source toolkit for MPEG transport stream analysis, manipulation, and monitoring, used in digital television and video streaming.