Telecom & Wireless Papers Monitor (arXiv)

Get a clean, structured, deduplicated feed of new public research papers in wireless communications, signal processing, networking, information theory and systems — straight from the official arXiv API. Run it once for a literature snapshot, or schedule it daily or weekly to get only the papers you haven't seen yet.

Use cases

• R&D and standards teams: a weekly digest of new papers on topics you track (CSI feedback, ISAC, RIS, O-RAN, 6G).
• Literature reviews: export every paper in a category and date range to JSON, CSV or Excel in one run.
• AI / RAG pipelines: clean title + abstract + metadata records, ready to embed or feed to an LLM.
• Alerts: schedule the Actor and connect it to Slack, email, Google Sheets or webhooks with Apify integrations.

What it does

• Searches arXiv by category (default cs.IT, eess.SP, cs.NI, eess.SY) and optional keywords (e.g. massive MIMO, CSI feedback, O-RAN, ISAC, 6G) with AND/OR logic.
• Primary category only (on by default): keeps papers whose main category is one you selected, dropping papers that are only cross-listed into it (e.g. robotics papers cross-listed to eess.SY).
• Filters by a date window (last N days, or from/to dates).
• Outputs one tidy JSON record per paper: arXiv ID, version, title, authors, abstract, categories, dates, PDF and abstract links, DOI, journal reference, comments.
• Monitoring mode: remembers seen arXiv IDs in a named key-value store, so scheduled runs output (and bill) only new papers. Optionally treats new versions (v2, v3…) as new.
• Polite by design: official API only, ≥3 s between requests, paginated, identifying User-Agent, following the arXiv API Terms of Use. Only public metadata is collected; PDFs are not downloaded.

How to use

1. Pick your categories (the defaults cover wireless/telecom) and optionally add keywords.
2. Set the date window (Last N days) and Max records.
3. Click Start. Results appear in the Output tab and can be downloaded as JSON, CSV, Excel or via API.
4. For monitoring, tick Only output papers not seen in previous runs and add a Schedule (e.g. daily).

Input

Field	Default	Description
categories	["cs.IT","eess.SP","cs.NI","eess.SY"]	arXiv categories, OR-ed. cs.LG is very large; add it together with keywords (e.g. wireless, MIMO, channel)
primaryCategoryOnly	true	Keep only papers whose primary category is selected (aliases like cs.IT/math.IT, eess.SY/cs.SY count). Filtered papers are not output, not charged and not remembered
keywords	[]	Optional phrases matched in title/abstract
keywordsOperator	OR	OR = any keyword, AND = all keywords
keywordFields	title_abstract	title_abstract, title, or all
lastNDays	7	Date window; ignored if dateFrom/dateTo set; 0 = no date filter
dateFrom / dateTo	–	YYYY-MM-DD (UTC, inclusive)
dateField	submittedDate	or lastUpdatedDate
maxRecords	100	Max papers output (and charged) per run
sortBy / sortOrder	submittedDate / descending	also lastUpdatedDate, relevance
onlyNew	false	Skip papers seen in previous runs (free)
includeNewVersions	false	With onlyNew, output a paper again when a higher version appears
seenStoreName	telecom-papers-seen	Named key-value store for seen IDs; use a different name per monitor
maxScannedRecords	2000	With onlyNew/primaryCategoryOnly, cap on arXiv results scanned
contactEmail	–	Optional override of the contact in the User-Agent sent to arXiv

Example: a weekly monitor for AI-native air-interface papers.

{
  "categories": ["cs.IT", "eess.SP", "cs.NI"],
  "keywords": ["CSI feedback", "AI-native", "neural receiver", "beam management"],
  "keywordsOperator": "OR",
  "lastNDays": 8,
  "maxRecords": 500,
  "onlyNew": true,
  "seenStoreName": "ai-air-interface-monitor"
}

Output example

{
  "source": "arxiv",
  "arxiv_id": "2610.37424",
  "version": 1,
  "title": "Privacy-Preserving Over-the-Air Federated Learning over Cell-Free Massive MIMO System with User-Centric Clustering",
  "authors": [{"name": "Manobendu Sarker", "affiliations": []}, {"name": "Soumaya Cherkaoui", "affiliations": []}],
  "abstract": "This paper investigates cross-layer design for differentially private federated learning over cell-free massive MIMO networks ...",
  "categories": ["cs.IT", "eess.SY"],
  "primary_category": "cs.IT",
  "published": "2026-10-27T15:37:21Z",
  "updated": "2026-10-27T15:37:21Z",
  "pdf_url": "https://arxiv.org/pdf/2610.37424v1",
  "abs_url": "https://arxiv.org/abs/2610.37424v1",
  "doi": null,
  "journal_ref": null,
  "comments": null,
  "topics": null, "tasks": null, "methods": null, "datasets": null, "metrics": null,
  "llm_enriched": false,
  "is_new": true
}

topics, tasks, methods, datasets, metrics and llm_enriched are reserved for a possible future version and are always null / false in this version. Each run also saves a RUN_SUMMARY record (output/scanned/skipped counts, stop reason, the exact arXiv query) to its default key-value store.

Pricing

$2 per 1,000 papers ($0.002 per output paper, event paper-record). Papers skipped as already seen, or filtered out by Primary category only, are free. The run stops cleanly at your maximum charge limit, so you never pay more than you set. Example: a daily monitor that outputs ~40 new papers costs about $0.08 per run.

Tips

• arXiv announces new submissions with a delay (typically ~1 business day, none on weekends), so the most recent day of a window is usually still empty. For daily schedules use lastNDays ≥ 3 with onlyNew on. The overlap costs nothing because already-seen papers are not charged.
• Typical volume (checked Oct 2026, default categories, 7 days): ~385 papers incl. cross-lists, ~272 with Primary category only. cs.IT alone ≈ 137/week incl. cross-lists.
• Each arXiv page (≤100 papers) takes ≥3 s because of the rate limit, so 1,000 papers take about 30 s or more.
• Abstracts may contain LaTeX ($...$); it is passed through unchanged.
• Known limitation: an old paper newly cross-listed into a category keeps its original submission date, so a date-window search doesn't catch it (≈1 in 137 in our check).

FAQ

Is this allowed? Yes. The Actor uses arXiv's official, public API within its published rate limits and only collects public metadata (titles, abstracts, authors, links). It does not download PDFs.

Why are there fewer papers than in the category listing? By default, Primary category only drops papers that are only cross-listed into your categories. Turn it off to include them.

Can I run several monitors? Yes. Give each one its own seenStoreName so their "seen" lists don't mix.

Roadmap (planned, not available yet)

• IETF drafts/RFCs and 3GPP specification metadata (metadata, abstract and links only).
• Optional AI structuring of papers into the reserved fields above.

Data & attribution

Metadata is retrieved from the arXiv API. Thank you to arXiv for use of its open access interoperability. Paper content remains under each paper's own license; this Actor does not download or redistribute PDFs.