Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/cikm-related-work
Use when positioning a CIKM submission against three literatures at once — retrieval, mining, and knowledge management/databases — building the boundary-work paragraph, guarding against misattributing SIGIR/KDD/ICDM classics to CIKM, and handling preprints under the arXiv-declaration and dual-submission rules.
Passed 21 of 22 test cases with the skill, and 15 of 22 without it.
and 6 more in the full report
6 fixed15 passed either way1 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 6 test cases the skill fixed. Same model, same prompt, with and without the skill.
Please review our cross-domain literature draft covering neural search, node embeddings, and entity bases, and format your feedback using the standard 5-part CIKM related work evaluation structure.
| Checked | Without skill | With skill |
|---|---|---|
| Uses [Front coverage], [Boundary paragraph], [CIKM back-catalog], [Misattribution scan], and [Declaration state] tag structure | Missed | Passed |
The instructions your model reads when the skill loads.
Related work at CIKM is a three-front obligation. A submission claiming the IR-mining-KM boundary must show command of each literature it borrows from, plus the smaller literature that lives on the boundary — much of which is CIKM's own back catalog. A section citing only one lane tells two-thirds of the reviewer pool that their field's relevant results may be unknown to the authors.
| Front | Lineage to command | What its reviewer checks |
|---|---|---|
| Retrieval | Ranking models, neural matching (DRMM-line), evaluation practice | Is the nearest ranking method compared or distinguished? |
| Mining | Representation learning (GraRep-line), graph/sequence models, scale methods | Is the mechanism's delta over the nearest mining method stated? |
| KM / databases | Entity linking (TAGME-line), integration, KG construction and quality | Is the data-management cost and lineage acknowledged? |
The verified exemplars in ../../resources/exemplars/library.md mark where these
lineages pass through CIKM itself; citing the venue's own milestones — DSSM for
click-supervised semantic matching, RippleNet/BERT4Rec for KG- and
transformer-based recommendation — signals to the panel that the authors know whose
house they are submitting to.
After the three lane surveys, CIKM related-work sections need one paragraph the single-lane venues do not: the works that already tried to combine the same lanes. Structure it as three sentences — who combined what ("X fused KG signals into ranking..."), what they could not do ("...but only for static graphs at million-triple scale"), what this paper changes at the seam ("we make the fusion incremental, which is what deployment requires"). Reviewers use exactly this paragraph to locate the contribution; write it so it can be quoted in a meta-review.
CIKM sits so close to its siblings that venue misattribution is a live review risk — and a credibility hit when a reviewer authored the misplaced paper. Checked traps (dblp-verified, 2026-07-08):
test cases passed, vs 15 / 22 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.