Individual skill
recsys-reproducibility
Use when strengthening the reproducibility of an ACM RecSys paper or preparing a RecSys Reproducibility Track submission — pinning dataset versions and splits, tuning baselines under an equal budget, reporting seeds and variance, avoiding sampled-metric distortion, and structuring a reproduction study with honest divergence analysis.
RecSys-Skills/skills/recsys-reproducibility/SKILL.md
Source commit 932eb23331721b16e967c59f093781be1928b498
What was checked
- Matches the observed repository HEAD
- Yes
- Scanner rules are current
- Yes
- Repository scan complete
- No
- Skill scan complete
- Yes
- Observed behavior
- Network or broad shell
- Last checked
- Nov 9, 2026
These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.
Install this skill
Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.
An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.
Repository context
- Repository stars (not skill ratings)
- 1,405
- Repository last updated
- Sep 27, 2026
- License
- MIT
Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.
- academic-writing Repository topic
- agent-skills Repository topic
- journal Repository topic
- academic-research Repository topic
- ai-agents Repository topic
- anthropic Repository topic
- awesome-list Repository topic
- claude Repository topic
- claude-code Repository topic
- economics Repository topic
- finance Derived category
- research Derived category
Files and excerpt
File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.
Browse files
SKILL.mdcore · 4,693 bytes
Read excerpt
# RecSys Reproducibility
Reproducibility is unusually load-bearing at RecSys for two reasons. First, the venue runs a
dedicated **Reproducibility Track** for papers that repeat, refute, or re-scope prior results and
for datasets and frameworks that enable future reproduction. Second, the field's own literature
documented how often reported recommender gains fail to survive a fair re-evaluation, so reviewers
of *regular* papers read reproducibility signals as a proxy for whether the gains are real.
## Why recommender results drift: the usual suspects
| Drift source | Typical symptom | Pin it by |
|---|---|---|
| Dataset version / filtering | "MovieLens" numbers differ across papers | Exact version id, k-core filter, and checksums |
| Split protocol | Session results inflated | Temporal or leave-one-last split, not random; document it |