← 최신 프로덕트

Entropy-Guided Loop

How to make small models reason. A notebook that compares a reasoning model x a non reasoning model that runs a loop using logprobs found uncertainty. TLDR: A small, vendor-agnostic inference loop that turns token logprobs/perplexity/entropy into an extra pass and reasoning for LLMs.

오픈소스

25 스타 · GitHub외부 사이트

github.com 방문외부 사이트

반응 (1)

비슷한 프로덕트

설명이 비슷한 순

  1. An educational Local Qwen3 LLM Inference project written in Rust

    github.com

    An educational Rust project for exporting and running inference on Qwen3 LLM family

    Hacker News15점

  2. PageIndex for Reasoning-Based RAG

    vectifyai.notion.site

    Large Language Models (LLMs) are powerful tools for language understanding and generation, but they suffer from a fundamental limitation: the context window. Most LLMs can only process a limited number of tokens at a time. For instance, Claude does not allow uploading very long documents in full, which makes tasks like long document analysis or question answering particularly challenging.

    Hacker News6점

  3. Flow Matching model inference in C

    github.com

    C-Flow Matching inference

    Hacker News5점

  4. I made a state machine framework for guided conversations with LLMs

    github.com

    GoalChain for goal-orientated LLM conversation flows

    Hacker News19점

  5. Fixing LLM memory degradation in long coding sessions

    github.com

    Empirical proof of SOTA LLM (GPT-5/Gemini-Pro/Claude-Pro) context saturation in complex engineering. Contains the "Misuraca Protocol" for deterministic logical segmentation to prevent entropy drift.

    Hacker News5점