Empezar
Awesome AI Research Writing - Online Skill

Awesome AI Research Writing - Online Skill

Stop copy-pasting prompts from a doc. Research Writing Assistant routes your academic writing task to the right expert prompt automatically — translation (ZH↔EN, LaTeX & Word), polishing, compression, expansion, de-AI, figure generation, caption writing, experiment analysis, mock review, and chart recommendation. Built from workflows used at MSRA, ByteDance Seed, and top CS PhD programs.
#Escritura#Investigación
Valoración
Se necesitan más valoraciones
Vendidos
0
Cómo se usa
Ejecutar en Capafy
También en aplicaciones externas
Proporcionado por el publisher
Claude Sonnet 4.6

Research Writing Assistant

Academic writing micro-tasks for researchers targeting NeurIPS, ICML, ICLR, ACL, and AAAI — without spending time finding or tuning the right prompt.

Describe what you need in plain language. The assistant routes to the correct expert template automatically.


What It Covers

Translation

  • Chinese → English (LaTeX) — Academic-grade translation with LaTeX special character escaping, present-tense enforcement, and zero AI-flavored filler.
  • Chinese → English (Word) — Same quality, plain text output for direct paste into Word. No Markdown contamination.
  • English → LaTeX → Chinese — Literal translation for comprehension; strips \cite{}, \ref{}, and math commands into readable text.

Polishing

  • English polish (LaTeX) — Deep rewrite targeting top-venue standards: sentence structure, word choice, grammar, zero em-dashes, no itemization.
  • Chinese polish (Word) — Conservative editing. Only intervenes when there is a genuine flaw — never rewrites clean prose for the sake of it.

Length Control

  • Compress — Trim 5–15 words without losing any technical detail. Clause-to-phrase conversions, redundancy removal.
  • Expand — Add 5–15 words by surfacing implicit logic, cause-effect relationships, and missing connectives. No padding.

De-AI

  • LaTeX — Removes "it is worth noting that," overused em-dashes, triple-item stacking, "showcase/delve/leverage," and uniform sentence rhythms.
  • Chinese (Word) — Removes "赋能/助力/深度融合," symmetric paragraph structures, and excessive intensifiers.

Figures & Captions

  • Framework figure — Generates a flat-vector academic illustration from your abstract + methods description. DeepMind/OpenAI paper aesthetic. All labels in English.
  • Figure caption — 2–4 sentences, ≤80 words, starts with "Overview of…" or "Comparison of…" — never "This figure shows."
  • Table caption — Title Case for noun phrases, Sentence case for full sentences. Standard expressions: "Comparison with," "Ablation study on," "Results on."

Analysis & Review

  • Experiment analysis — Turns raw numbers into \paragraph{}-structured LaTeX analysis. Trend-first, no data fabrication, no boilerplate "A achieves B."
  • Chart recommendation — Picks from 19 standard academic chart types matched to your data structure. Flags when error bars or log scale are needed.
  • Mock review — Full reviewer report: Summary, Strengths, Weaknesses (Critical), Rating (1–10), plus strategic revision advice with actionable experiment suggestions.
  • Logic check — Sentence-level audit for unsupported claims, correlation-causation confusion, internal contradictions, and scope overgeneralization.

Scope

Supports: LaTeX and Word workflows, Chinese and English input.

Not for: writing full papers from scratch, fetching citations from the web, or generating experimental data.