paper-radar
Paper Radar finds papers linked to major AI companies and research institutes across any selected date range.
Every report separates papers led by a lab from papers that only include a lab somewhere in the author list.
Affiliation data is missing
arXiv does not reliably store author affiliations.
The <arxiv:affiliation> field has roughly 1 percent coverage. Semantic Scholar reaches about 7 percent. OpenAlex returns no affiliation data for these preprints.
Paper Radar reads the author block from each paper and matches affiliations through email domains, ROR records, official name variants, and a controlled lab alias table.
| Data source | Available information |
|---|---|
| arXiv API | Paper metadata and submission dates |
| Paper HTML | Authors, superscripts, affiliations, and email domains |
| ROR | Verified organization identities and official name variants |
| Paper Radar | Lab attribution with lead authorship separated from byline appearances |
How it works
| Step | What happens |
|---|---|
| 0 · Listing | arXiv API: category × submittedDate, deduped by ID |
| 1 · Full text | arxiv.org/html/{id}, Range-request the first 90KB |
| 2 · Structure | Map (author, superscripts) against (superscript, affiliation) |
| 3 · Entity | Email domain, then ROR name variants, then the lab alias table |
| 4 · Grading | Lead = the first author's institution |
| 5 · Supplements | Apple RSS, MSR embedded JSON, arXiv team-name queries |
| 6 · Report | Dedupe, write markdown, keep the evidence for every match |
The report gives every lab two columns, and the gap between them is the whole point:
| Company | Total hits | Of which lead |
|---|---|---|
| Microsoft | 31 | 18 |
| Meta | 9 | 9 |
| Apple | 8 | 0 |
Coverage: 28 labs — Google, Microsoft, Meta, Apple, NVIDIA, OpenAI, Anthropic, Alibaba, Tencent, ByteDance, DeepSeek, and more. Each is anchored to a real ROR ID, carrying 98 official name variants between them, Chinese included. Every match records what it matched on and the raw text it matched, so you can check any row by eye.
Get started
Paste this into Claude Code, Codex, or anything else with a shell:
Install the paper-radar skill from https://github.com/tigerless-labs/paper-radarThat is the whole install. Then ask:
/paper-radar what did big tech publish on arXiv in the last two weeks?
has Xiaomi published anything on self-evolving agents?Or run it yourself, no agent involved:
python3 skills/paper-radar/scripts/paper_radar.py --days 14A two-week window (~3000 papers) takes about three minutes. Pure stdlib, nothing to build.
More in the works — built in the open, shipped fast.
Explore Tigerless Labs