인터페이스는 현지화되어 있으며, Skill 소개와 설명 문서는 선택한 언어로 AI가 자동 번역합니다. 코드·명령어·경로·전문 용어는 원문을 유지하며 원작자 문서가 우선합니다.
Skills 둘러보기
데이터 및 연구 · SKILL.md

nature-literature-pipeline

nature-literature-pipeline

Build a repeatable literature discovery, scoring, reading, and archiving pipeline for RL, MARL, LLM, and LLM Agent research. Use for daily/weekly paper tracking, topic watchlists, benchmark monitoring, and research-note automation.

Independent GitHub snapshot · Automated classification · Original documentation governs

Skill 문서

GitHub

name: nature-literature-pipeline description: Build a repeatable literature discovery, scoring, reading, and archiving pipeline for RL, MARL, LLM, and LLM Agent research. Use for daily/weekly paper tracking, topic watchlists, benchmark monitoring, and research-note automation. version: 2.0.0-agent-rl

Literature pipeline skill

Workflow

  1. Define scope: topic, venues, keywords, authors, benchmarks, arXiv categories.
  2. Define sources: arXiv, OpenReview, Papers with Code, venue proceedings, official benchmark repos.
  3. Score papers using six axes: relevance, novelty, evidence, benchmark match, reproducibility, idea value.
  4. Produce a triage table: read now, skim, archive, ignore.
  5. Generate one-page notes for high-priority papers.
  6. Maintain a local index with BibTeX, code links and relation to current projects.

Output format

Pipeline design:
Sources:
Search queries:
Scoring rubric:
Weekly triage table:
Archive structure:

Red lines

Do not present unverified paper metadata as final. Do not automate downloads from restricted sources.

저장한 Skills