← all publishers

WooyoohL

@wooyoohl source repo

Report this profile Claim this profile

• 14 skills
• 0 followers
• 9 hours ago last updated

1. ▌
   Candidate Claim Set · wooyoohl

   Use after problem-formulation to convert clearly defined research problems into 2–4 falsifiable candidate claims without prematurely committing to one paper shape.
   0 installs
2. ▌
   Entry Source Intake · wooyoohl

   Use at the beginning of an AI/ML research-idea session to identify the user's entry state, available materials, target area, and resource constraints before invoking deeper modules.
   0 installs
3. ▌
   Problem Formulation · wooyoohl

   Use after research-signal-mining and before candidate-claim-set to define the actual research problem an AI/ML paper idea solves, explains, measures, or refutes. Separates signal, problem, claim, and method so later novelty, scope, and experiment decisions are grounded.
   0 installs
4. ▌
   Reviewer Risk Check · wooyoohl

   Use late in the research pipeline to identify the top fatal reviewer risks for a residual claim and evidence plan, without overfitting to imagined reviewers.
   0 installs
5. ▌
   Research Move Router · wooyoohl

   Use to provisionally classify each candidate claim into the most plausible paper contribution type without forcing target reframing everywhere.
   0 installs
6. ▌
   Paper Shape Generator · wooyoohl

   Use at the end of the pipeline to turn a residual claim, scoped contribution, evidence plan, and reviewer risks into a coherent paper pitch and execution plan.
   0 installs
7. ▌
   Paper Unit Scope Gate · wooyoohl

   Use after prior-work triangulation to judge whether the residual contribution is too thin, too broad, or well-scoped enough to form one coherent AI/ML paper.
   0 installs
8. ▌
   Research Signal Mining · wooyoohl

   Use to extract idea seeds from a field map, papers, failed experiments, reviewer comments, benchmark artifacts, or unexplained empirical phenomena.
   0 installs
9. ▌
   AI Research Orchestrator · wooyoohl bundle

   Entry point for modular AI/ML research idea discovery. Use when the user wants to find, evaluate, sharpen, or execute paper-oriented AI research ideas. Routes through literature bootstrapping, signal mining, problem formulation, claim formation, prior-work triangulation, scope gating, decision-value experiments, reviewer risk checks, and paper-shape planning while maintaining research_state artifacts.
   0 installs
10. ▌
    Research Target Reframing · wooyoohl

    Use when analyzing research papers, research ideas, ML/AI methods, or project proposals to identify the core problem representation, hidden complexity, simplifying target, first-principles reformulation, ablation plan, and “why not simpler?” research taste. 适用于论文拆解、研究品味蒸馏、idea 评估、方法重构、实验设计、找 surprise/不对劲现象。
    0 installs
11. ▌
    Claim Prior Work Triangulation · wooyoohl

    Use to collide a candidate claim with nearest prior work and extract the residual contribution that remains after accounting for existing papers.
    0 installs
12. ▌
    Decision Value Experiment Planner · wooyoohl

    Use to design AI/ML experiments that answer explicit hypotheses and support research decisions, avoiding low-information-gain exhaustive experiments and invalid cheap proxies.
    0 installs
13. ▌
    Research State Artifact Management · wooyoohl bundle

    Use throughout the research pipeline to checkpoint progress into documents, manage project directories, preserve decisions, and prevent context drift caused by long conversations.
    0 installs
14. ▌
    Representative Literature Bootstrap · wooyoohl

    Use before deep idea generation to build a working map of a target AI/ML field by reading representative papers, especially introductions, related work, formulations, experiments, limitations, and reviews.
    0 installs