Workflows / 工作流

Workflow leaderboard

工作流实时榜

The current best-verified tool combo for common workflows, grouped by domain and task. Research packs are live first; more domains fill as they are curated. Live version with interactive detail: https://trendsentinel.net/discover/workflows

Research / 科研 (18)

Transcription / 转写 (3)

• Transcribe and diarize research interviews / 研究访谈转写与说话人分离

  Turn recorded interviews into timestamped, speaker-labelled transcripts while preserving a reproducible path from audio to JSON and subtitle

• Turn interview transcripts into coded themes / 访谈转写稿主题编码

  Produce a traceable first-pass qualitative codebook and coded excerpts from interview transcripts without losing links to speaker turns.

• Batch-transcribe lectures into reviewable subtitles / 批量转写讲座并生成可复核字幕

  Convert a folder of lectures or field recordings into timestamped subtitles, with a human correction pass and traceable source files.

PDF & document processing / 文献处理 (3)

• OCR, translate and structure research PDFs / 科研 PDF OCR、翻译与结构化

  Convert scanned or mixed-language research PDFs into searchable PDFs plus structure-preserving translated Markdown and JSON artifacts.

• Extract PDF tables into analysis-ready data / PDF 表格转分析数据

  Extract tables from research PDFs into CSV and JSON with page provenance and validation checks before analysis.

• Convert scholarly PDFs to auditable Markdown / 将学术 PDF 转为可审计 Markdown

  Extract born-digital or scanned papers into structured Markdown while retaining page-level artifacts for verification.

Literature review / 文献综述 (3)

• Screen records for a systematic review / 系统综述文献筛选

  Deduplicate, prioritize and screen a literature-search export while preserving reviewer decisions and an auditable exclusion trail.

• Expand a literature review through citation networks / 通过引文网络扩展文献综述

  Grow a defensible candidate-paper set from seed papers using backward, forward and similarity-based discovery, then deduplicate it in a refe

• Build cautious local literature retrieval and RAG / 搭建谨慎型本地文献检索与 RAG

  Retrieve page-labelled passages from a local paper collection and generate answers that separate cited evidence from uncertainty.

Academic writing / 学术写作 (2)

• Write a citation-safe academic manuscript / 可追溯引用的学术写作

  Draft and compile a manuscript with stable citation keys, reproducible bibliography exports and explicit checks for unsupported claims.

• Compare free and paid academic English polishing / 对比免费与付费学术英语润色

  Compare local grammar checking with paid academic-language editing on the same blinded sample without allowing either tool to alter scientif

Data analysis / 数据分析 (3)

• Build a reproducible Python data analysis / 可复现的 Python 数据分析

  Turn raw tabular files into validated Parquet data, scripted analysis and a reproducible environment with a clear raw-to-result lineage.

• Build a reproducible R statistical report / 可复现的 R 统计报告

  Combine statistical code, narrative, tables and figures in one versioned report with a restorable R package environment.

• Create a reproducible R and Python analysis environment / 创建可复现的 R 与 Python 分析环境

  Run R and Python in one locked project, render a single analysis report and archive the environment and input hashes.

Figures & plotting / 绘图 (2)

• Create publication-ready scientific figures / 制作可投稿的科研图

  Generate reproducible data figures, edit vector annotations safely and export journal-ready files with documented dimensions and fonts.

• Produce a research poster and academic slides / 制作科研海报与学术演示文稿

  Reuse one verified claim and figure inventory to produce a print-ready poster and reproducible conference slides.

Reference management / 文献管理 (2)

• Build a reference-library-to-citation pipeline / 从文献库到引用的管理流程

  Maintain one deduplicated reference library with stable citation keys, attachment backups and reproducible exports for writing tools.

• Check reference format and citation round trips / 检查参考文献格式与引文往返一致性

  Apply a journal CSL style reproducibly, surface unresolved citation keys and verify that in-text citations match the bibliography.

General / 通用 (6)

Coding assistant / 代码助手 (2)

• Use a coding assistant with research guardrails / 带科研护栏的代码助手

  Accelerate research-code changes while requiring tests, data-lineage checks and human review before accepting generated patches.

• Automate a research pipeline with provenance / 自动化科研流程并保留来源记录

  Replace an undocumented sequence of analysis commands with a restartable workflow that records inputs, outputs and software configuration.

Data collection / 数据收集 (2)

• Collect research surveys with a privacy-aware workflow / 用隐私审慎流程收集科研问卷

  Design, pilot, collect and export survey responses with stable variable names, consent controls and a frozen analysis dictionary.

• Collect reproducible behavioral experiment data / 采集可复现的行为实验数据

  Build and pilot a behavioral experiment with versioned stimuli, explicit randomization and row-level provenance before collecting participan

Corpus annotation / 语料标注 (2)

• Annotate a corpus with an agreement gate / 带一致性门槛的语料标注

  Run double-blind text classification, quantify annotator agreement and adjudicate disagreements without overwriting raw labels.

• Annotate named entities with boundary adjudication / 标注命名实体并裁决边界分歧

  Create span-level entity labels with explicit boundary rules, independent annotation and a lossless adjudication record.