Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/acmmm-related-work
Use when building or auditing the related-work section of an ACM MM (ACM Multimedia) paper — covering the multimedia literature spread across vision, audio/speech, language, HCI/QoE, and systems, handling arXiv-speed concurrency, keeping citations double-blind, and verifying that cited "ACM MM papers" are ACM MM and not ICMR, MMSys, CVPR, or TOMM.
Passed 17 of 22 test cases with the skill, and 12 of 22 without it.
and 3 more in the full report
5 fixed12 passed either way5 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.
I need a complete diagnostic audit of my ACM MM literature review draft across all required audit dimensions. Output the report using the standard diagnostic format.
| Checked | Without skill | With skill |
|---|---|---|
| Formats the output block using all six required bracketed tags including [Shelf coverage], [Fusion lineage], [Deltas], [Concurrency], [Blindness], and [Venue hygiene]. | Missed | Passed |
The instructions your model reads when the skill loads.
Use this to position an ACM Multimedia paper against a literature that is unusually spread out: a cross-modal contribution touches several single-modality communities plus the multimedia venues that combine them.
A strong ACM MM related-work section shows command of all shelves the contribution touches, not just the author's home community:
| Task | What to do | Failure it prevents |
|---|---|---|
| Cover each modality you use | Cite the current best single-modality work per stream | "You ignored the vision literature" |
| Cite the fusion lineage | Trace the cross-modal line your method extends | "No delta over existing fusion" |
| Verify the venue string | Confirm each "ACM MM" cite on dblp conf/mm | Misattributing an ICMR/CVPR paper to ACM MM |
| Handle concurrency | Note contemporaneous arXiv work honestly | "You missed / overclaimed novelty vs. X" |
| Keep it blind | Cite your own prior work in third person | Double-blind violation |
Do not enumerate. For each closest neighbor, state in one clause what it did and in one clause what your paper adds — and make sure the delta is cross-modal, since "we swap a better encoder" is a single-modality delta a reviewer will discount.
[Neighbor] Late-fusion audio-visual highlight scoring (VenueYear).
[Their move] Average per-modality scores.
[Our delta] Score the timing DISAGREEMENT between streams — a signal averaging cannot represent.
test cases passed, vs 12 / 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.