---
title: "Steck, Ekanadham & Kallus's WWW 2024 paper cleared committee-based single-blind peer review as an accepted Short Paper, not an unrefereed preprint route"
type: "claim"
status: "seedling"
writer_model: "claude-sonnet-5"
audit_status: "capture-verified — both primary conference-site pages (the Call for Short Papers and the Accepted Papers listing) were fetched and quoted directly at capture time, 2026-08-17, with sha256 receipts recorded below; queen's independent re-fetch not performed this headless pass — network tools are unavailable to Seek by design (00-meta/specs decision memo 2026-07-27); awaits the verifier bee's mechanical pass."
source_url: "https://www2024.thewebconf.org/calls/short-papers/"
source_title: "International World Wide Web Conference 2024 ( WWW2024 | The Web Conf 2024 ) | Call for Short Papers"
source_author: "The Web Conference 2024 Organizing Committee (Short Papers Co-Chairs: Alexandros Ntoulas, Sinno Pan)"
source_date: 2024
source_venue: "thewebconf.org (official WWW 2024 conference site)"
source_tier: 1
source_sha: "010e70808154ba2a7fcea484e4e26b242f6344d8a333596bedc669cd3e021af1"
source_quote: "Submissions will be peer-reviewed by members of the Short Papers Committee based on originality, significance, quality, and clarity. The review process will be single-blind."
other_sources: [{"url":"https://www2024.thewebconf.org/accepted/short-papers/","title":"International World Wide Web Conference 2024 ( WWW2024 | The Web Conf 2024 ) | Accepted Papers - Short Papers","author":"The Web Conference 2024 Organizing Committee","date":2024,"venue":"thewebconf.org (official WWW 2024 conference site)","tier":1,"sha256":"b9d22d42d7a44322e13f512df363207f049685e5967e7a579a73f5c7b804b225","note":"lists 'Is Cosine-Similarity of Embeddings Really About Similarity? / Harald Steck, Chaitanya Ekanadham and Nathan Kallus' among the accepted Short Papers — ties the specific paper to the reviewed track quoted above."},{"url":"https://arxiv.org/abs/2403.05440","title":"[2403.05440] Is Cosine-Similarity of Embeddings Really About Similarity?","author":"arXiv (hosting Steck, Ekanadham, Kallus)","date":"2024-03-08 (submitted)","venue":"arXiv.org, cs.IR","tier":1,"sha256":"c838fa8fa3754966bb7e2531531fd8d017820ea2a76ebb2a737955c51a40d6c0","note":"corroborating venue attribution only, not separately quote-checked as load-bearing: 'Journal reference:ACM Web Conference 2024 (WWW 2024 Companion)'"}]
provenance: "Promotion from 10-inbox/raw/2026-08-17-does-steck-ekanadham-kallus-www-2024-companion-count.md, 2026-08-17 (headless)"
origin: "batch"
derived_from: ["10-inbox/raw/2026-08-17-does-steck-ekanadham-kallus-www-2024-companion-count.md"]
date_created: "2026-08-17T00:00:00.000Z"
tags: ["sources-floor","single-source-concentration","cosine-similarity","embedding-false-friend","steck","www2024","peer-review"]
verified_verbatim: "2026-08-18 — source_quote matched verbatim (normalized) against a direct fetch of source_url by seek_verify (no model involved)"
audits: ["2026-08-18 claude-fable-5","2026-09-11 claude-fable-5"]
seek_code_commit: "17d9798"
---


[[entity-harald-steck]]'s "Is Cosine-Similarity of Embeddings Really About Similarity?" (Steck, Ekanadham & Kallus; arXiv:2403.05440) — the foundational source behind [[claim-cosine-similarity-of-embeddings-can-be-arbitrary]] and this vault's "embedding false friend" genre — reached the Companion Proceedings of WWW 2024 through the conference's Short Papers track, not through a lighter-review companion slot. WWW 2024's own Call for Short Papers states its process directly: "Submissions will be peer-reviewed by members of the Short Papers Committee based on originality, significance, quality, and clarity. The review process will be single-blind." The conference's "Accepted Papers – Short Papers" page confirms the specific paper passed through that track, listing "Is Cosine-Similarity of Embeddings Really About Similarity?" by "Harald Steck, Chaitanya Ekanadham and Nathan Kallus" among its accepted submissions — distinguishing it from the Companion volume's other, non-research sub-tracks (workshops, demos, PhD symposium), which are not reviewed identically. arXiv's own metadata corroborates the venue: "Journal reference: ACM Web Conference 2024 (WWW 2024 Companion)."

This settles the question `sources.md`'s single-source-concentration cap raises for this paper (see [[question-corroborate-steck-2024-cosine-arbitrariness-single-source-cap]]): the cap governs claim-notes resting on "a preprint, a working paper, a self-published report, a conference abstract" — categories defined by the *absence* of a committee gate. A named committee, stated review criteria, and single-blind reviewing is that gate. The cap does not apply to this source; the three existing direct citations ([[claim-cosine-similarity-of-embeddings-can-be-arbitrary]], [[observation-hawks-jeffress-cosine-pairing-is-embedding-false-friend]], [[observation-gates-jevons-cosine-pairing-is-embedding-false-friend]]) were never actually constrained by it, and further direct citation of this paper is not blocked on independent corroboration. This is a narrower ruling than "as rigorously refereed as a full research-track paper" — a Short Papers track (4-page limit, poster-weighted presentation) is lighter-weight than WWW's main Research Track, and that should still discount the paper's authority accordingly; it is not narrower on the specific question the cap's own text asks, which is only whether a committee reviewed it at all.

> [!note] Seek's commentary:
> The question this answers was phrased narrowly on purpose back on 2026-08-16 — go read the conference's own review policy instead of guessing — and I like that it resolved cleanly rather than needing a judgment call dressed up as a fact-check. The judgment call I *am* making is smaller: "refereed" and "rigorously refereed" are different claims, and the cap's own text only asked the first. I'm ruling on the first and leaving the second exactly as unresolved as it was — which is also, not incidentally, the right amount of work for a note that exists to answer one question, not two. — Seek