惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

C
Check Point Blog
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
博客园 - 聂微东
月光博客
月光博客
博客园 - 司徒正美
爱范儿
爱范儿
aimingoo的专栏
aimingoo的专栏
量子位
Recent Announcements
Recent Announcements
V
V2EX
P
Proofpoint News Feed
小众软件
小众软件
云风的 BLOG
云风的 BLOG
腾讯CDC
宝玉的分享
宝玉的分享
Microsoft Azure Blog
Microsoft Azure Blog
大猫的无限游戏
大猫的无限游戏
Vercel News
Vercel News
The GitHub Blog
The GitHub Blog
A
About on SuperTechFans
B
Blog
博客园_首页
GbyAI
GbyAI
博客园 - Franky

Hacker News - Newest: "LLM"

GitHub - lechmazur/position_bias: A benchmark for testing whether LLM judges keep the same preference when two lightly edited versions of the same story are shown in opposite orders. Flex routing (EU and EFTA) Dark Factories: Retooling for LLM Velocity Ask HN: What would be the impact of a LLM output injection attack? GitHub - Oaklight/llm-rosetta: Production-ready LLM API translation layer for Python — bidirectional conversion between OpenAI, Anthropic & Google formats via hub-and-spoke IR. Optional API gateway. Streaming & non-streaming. Zero core deps. Contributions welcome! GitHub - browser-use/browser-harness: Self-healing browser harness that enables LLMs to complete any task. GitHub - moeen-mahmud/remen: Remen turns thoughts into something you can return to Analyzing 156 LLM Launch Posts on Hacker News ChatGPT vs Gemini vs Claude: The Best LLM Subscription You Should Buy GitHub - salaamalykum/quran-semantic-search: High-density RAG Semantic Search Engine & Quran Corpus (GEO/SEO Architecture) GitHub - NVIDIA/TensorRT-LLM: TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT LLM also contains components to create Python and C++ runtimes that orchestrate the inference execution in a performant way. The State of LLM Bug Bounties in 2026 Operational Readiness Criteria for Tool-Using LLM Agents Meshcore: Architecture for a Decentralized P2P LLM Inference Network How an LLM becomes more coherent as we train it GitHub - seetrex-ai/laimark GitHub - Jossifresben/BibCrit: AI-assited biblical textual criticism GitHub - wastedcode/memex: File system based wiki, maintained by Claude 99helpers.com GitHub - cliver-project/AITrigram GitHub - unbody-io/adapt: A self-evolving memory layer for AI agents. GitHub - hb20007/awesome-gen-ai-fails: A list of incidents where reliance on generative AI and LLMs resulted in harm to companies, individuals, or society GitHub - nevenkordic/localmind: Run any local LLM with persistent memory and context. CLI agent over Ollama with SQLite-backed hybrid recall. No cloud. Ask HN: What are the machine requirements for a LLM like Llama-3.1-8B? Faster LLM Inference via Sequential Monte Carlo grpo explained: group relative policy optimization for llm finetuning - cgft Stop comparing price per million tokens: the hidden LLM API costs · TensorZero Andrej Karpathy's LLM Wiki Is a Bad Idea GitHub - GG-QandV/mnemostroma: Offline RAM-first cognitive leer/coprocessor for AI agents and robotics. Solves "Context Abandonment" with 20-80ms latency using a dual-thread biomimetic memory architecture (ONNX + SQLite WAL). mempalace/agent at agent · skorotkiewicz/mempalace
GitHub - fu5ha/pi-treebase: Interactive-rebase style tree...
MrBuddyCasin · 2026-05-12 · via Hacker News - Newest: "LLM"

A session history management tool that combines the functionality of the base /tree command with something similar to git rebase --interactive.

demo gif

Installation

pi install npm:@grayolson/pi-treebase

Usage

First, use the /treebase and navigate with the normal tree session view to choose the destination to rebase history to.

Then, use an interactive editor of flattened history to select what to Pick (keep verbatim), summarize with Low or High importance, or Drop entirely.

A new synthetic history branch is then created from the root you selected, interspersing actual picked history entries with generated summaries.

Summaries are generated by a model with full context of the entire branch (except dropped entries), but with instructions to only record a summary of the relevant section of the entire linear history into each summary block so that summaries makes sense when interspersed with picked entries.

Limitations

Currently, the only supported operation is to "rebase" directly back in time to a parent node. Rebasing one branch past a junction on top of of another branch is not yet supported.

Hacks

  • Implementing this required (as far as I could tell) mutable access to pi's SessionManager, which is usually exposed as only ReadonlySessionManager to command handlers. I did a bad and just as-casted it to the writable version. I'm not sure exactly how bad the implications of this are.
  • Using a custom message types as a summary message doesn't behave super well with normal tree navigation, so I hackily pushed native branch summarization messages for this purpose by hooking a private method.

I will likely try to file issues/PR the ability to do these non-hackily once the bigrefactor lands.

Slop alert

Warning, this is moderately slop clanker code. It's been cleaned up a bit and I reviewed what I felt were the important parts, but do with that what you will.