// Use when the user wants to find related work, survey a research area, identify literature gaps, or discover open-source implementations. Triggers on phrases like "find papers on", "related work", "literature review", "what papers exist", "open source implementation", or "papers with code".
Same-epoch comparison of training runs across wandb, neptune, tensorboard, or mlflow. Aligns runs at the student's current step (never current-vs-final-of-baseline) and separates proxy metrics from downstream targets. Use when the user asks to compare runs, check if a run is improving, track lag against a baseline, rank experiments, or evaluate run-vs-run performance.
Evidence-before-action diagnosis of failing ML experiments. Probes the system before guessing causes, process list, dmesg, GPU stats, log scrollback, checkpoint state, then states a hypothesis as a hypothesis and runs a smoke before claiming a root cause. Use when the user asks why a run is failing, diverging, OOMing, hanging, slow, producing weird metrics, has crashed, or asks to debug, diagnose, troubleshoot, or investigate a training issue.
Pre-flight checklist for long-running ML training jobs covering config diff, run naming, path verification, monitoring setup, and restart-cleanup. Use when the user asks to launch, kick off, start, restart, or kill a training run, or mentions launching a multi-hour or multi-day GPU job (python train, accelerate launch, torchrun, deepspeed, sbatch, tmux training).
End-to-end paper reproduction from arxiv URL through smoke runs to replication experiments. Handles missing or partial official code, missing training scripts, missing hyperparameters, and private datasets via similar-public-dataset substitution. Use when the user asks to reproduce, implement, replicate, or re-run a paper from scratch, or pastes an arxiv URL with reproduction intent.
Use when the user wants to analyze dataset bias, create stratified samples, evaluate fairness, or plan dataset collection. Triggers on phrases like "dataset bias", "stratified sample", "class imbalance", "data distribution", "fairness analysis", or "ethical review".
Use when the user wants to design experiments, plan ablation studies, structure baselines, or create incremental evaluation strategies. Triggers on phrases like "design ablation", "plan experiment", "what experiments should I run", "baseline comparison", or "experiment matrix".
| name | literature-research |
| description | Use when the user wants to find related work, survey a research area, identify literature gaps, or discover open-source implementations. Triggers on phrases like "find papers on", "related work", "literature review", "what papers exist", "open source implementation", or "papers with code". |
You are helping a researcher conduct systematic literature research. Follow this methodology to ensure thorough, accurate coverage.
Before searching:
Use multiple search strategies in order:
From seed papers or initial results:
Organize found papers into a structured taxonomy:
| Paper | Year | Venue | Approach | Key Result | Code? | Relevance |
|-------|------|-------|----------|-----------|-------|-----------|
Group by methodology or approach type, not chronologically.
Map what exists vs. what's missing:
For each identified gap, verify it's real:
Rate confidence: HIGH (extensively searched, clearly missing), MEDIUM (searched but might have missed niche work), LOW (limited search, gap may exist elsewhere).
Produce a structured research landscape:
Every paper mentioned must have verified metadata: