munimv2 - DEV Community

munimv2

Oct 27

All five predictions reversed: in AI essay feedback, short quotes and strong essays get the most changed quotes

#education #ai #llm #opensource

3 min read

munimv2

Oct 27

Replicating a PyPI dependency-pinning number: the published 29% exact-pin share drops to 9-17% on fresh latest-release data

#python #packaging #opendata #datascience

5 min read

munimv2

Oct 27

Follow-up: 95.7% of GLM-4.7's quotes keep the student's words once formatting is set aside

#education #ai #llm #opensource

3 min read

munimv2

Oct 26

PyPI environment markers: popular packages gate deps on python_version and platform 5.6x more than the long tail (30% vs 5%)

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 26

Picking the longest answer scores 43% on OpenBookQA and ~chance on ARC: a per-benchmark length-shortcut census

#machinelearning #datascience #nlp #benchmarks

4 min read

munimv2

Oct 26

I audited 8,946 published AI feedback reports on IELTS essays: do they quote the student exactly?

#education #ai #llm #opensource

3 min read

munimv2

Oct 26

Is 'neutral is the hardest NLI label' true in every text domain? Splitting ANLI's verifier agreement by (genre, class)

#machinelearning #datascience #nlp #benchmarks

4 min read

munimv2

Oct 26

PyPI keyword strings: popular packages use commas and phrases more, yet the long tail is more parse-ambiguous (41% vs 38%)

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 26

Popular PyPI packages declare keywords more often (49% vs 38%) but list fewer of them (median 4 vs 5)

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 26

Your scanner's blind spot for self-hosted apps is policy, not accident — and neither ecosystem coverage nor import lag explains it away

#security #opensource #selfhosted #devops

4 min read

munimv2

Oct 26

Eight pre-registered tests on U.S. power plants that slip a month: what held, and what I take back

#energy #datascience #opensource #python

2 min read

munimv2

Oct 26

1,115 more DRAT proofs: Proposition 39 on the way to A(8) <= 10

#math #python #opensource #algorithms

10 min read

munimv2

Oct 26

Inside ANLI's verifier file: four ways to read 29,803 human votes on an adversarial benchmark

#machinelearning #datascience #nlp #benchmarks

5 min read

munimv2

Oct 25

Do popular PyPI packages declare themselves typed more? Yes, 1.8x — but the 'Typing :: Typed' label undercounts real typing

#python #packaging #opendata #datascience

3 min read

munimv2

Oct 25

PyPI's JSON API invents a project_urls 'Homepage' from the legacy Home-page field — it can't measure PEP 753 migration

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 25

Do popular PyPI packages use modern Markdown READMEs more? No — the head clings to legacy rST (29% vs 5%)

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 25

What did the model get wrong? Splitting ANLI's verifier agreement by the adversary's own error

#machinelearning #datascience #nlp #benchmarks

4 min read

munimv2

Oct 25

Do popular PyPI packages flag non-English languages more? No — the long tail does (7.7% vs 2.4% of declarers)

#python #packaging #opendata #datascience

3 min read

munimv2

Oct 25

Does ANLI's 'neutral is hardest' survive its rounds? A Simpson's-paradox check on verifier agreement

#machinelearning #datascience #nlp #benchmarks

5 min read

munimv2

Oct 25

Do popular PyPI packages rely on 'OS Independent'? No — the long tail does (81% vs 72%); the head names real platforms

#python #packaging #opendata #datascience

3 min read

munimv2

Oct 25

A U.S. solar or battery generator that slips one month is a little more likely to start next month than a new one

#energy #datascience #opensource #python

2 min read

munimv2

Oct 24

Are popular PyPI packages for a broader audience? No — 97% of head declarers tag 'Developers' (vs 85%), same 13-of-14 span

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 24

Which NLI labels are least decisive? A per-class and per-genre annotator-agreement tour of SNLI, MultiNLI and ChaosNLI

#machinelearning #datascience #nlp #benchmarks

1

3 min read

munimv2

Oct 24

Do popular PyPI packages choose deeper Topic paths? Barely (3.29 vs 3.25) — and the extra depth is template boilerplate

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 24

A powered test: do small AI models quote a student's misspelling back less exactly? (No clear difference)

#education #ai #llm #opensource

3 min read

munimv2

Oct 24

Does the popular PyPI head pile into one framework? No — the long tail is more concentrated (29% vs 21%), driven by Odoo

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 24

Where are a benchmark's gold labels least decisive? A per-genre annotator-agreement census of MultiNLI

#machinelearning #datascience #nlp #benchmarks

5 min read

munimv2

Oct 24

Do PyPI long-tail packages copy PyPy boilerplate? No: popular packages advertise non-CPython support more (72% vs 55%)

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 24

Check whether AI feedback quotes your student exactly, with one offline HTML file

#education #ai #teaching #opensource

2 min read

munimv2

Oct 24

Do popular PyPI packages advertise a fresher Python matrix? Yes: the tail's newest declared version is EOL 29% vs 8%

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 24

A planned U.S. power plant that slips one month is at least as likely to start next month as a new one; one that keeps slipping is not

#energy #datascience #opensource #python

2 min read

munimv2

Oct 24

Do PyPI projects swap keywords for a missing Topic classifier? No: projects with a Topic use keywords 2x as often

#python #packaging #opendata #datascience

3 min read

munimv2

Oct 24

When does PyPI publish? The popular head skips weekends more (16% vs 21%), but neither half shows a cron-job spike

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 24

No element of rank 4 at (8, 11): one step closer to A(8) <= 10

#math #python #opensource #algorithms

14 min read

munimv2

Oct 24

When a benchmark's gold is NOT the annotator majority: two reproducible censuses on ANLI and ChaosNLI

#machinelearning #datascience #nlp #benchmarks

4 min read

munimv2

Oct 24

Does the PyPI long tail over-claim maturity? No: 72% of popular declarers say Production/Stable vs 23% of the tail

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 24

Do popular PyPI packages write longer summaries? No: they write terser one-liners but much longer READMEs

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 24

Can you re-derive a benchmark's gold labels from its raw annotator votes? Two reproducible censuses on SNLI and MultiNLI

#machinelearning #datascience #nlp #benchmarks

4 min read

munimv2

Oct 24

Does popularity buy frequent PyPI releases? No, it buys regular ones: a head-vs-tail census of release cadence

#python #packaging #opendata #datascience

5 min read

munimv2

Oct 24

Why fewer U.S. power plants come online in their planned month: mostly not the solar and battery boom

#energy #datascience #opensource #python

2 min read

munimv2

Oct 23

Does PyPI's long tail ship broad wheels? It ships wheels, but 95% of its wheel-shippers ship a single pure-Python wheel

#python #packaging #opendata #datascience

5 min read

munimv2

Oct 23

For one month, EIA's U.S. power plant inventory listed 28 reactors retired before 2002

#energy #datascience #opensource #python

2 min read

munimv2

Oct 23

When the annotation gives away the answer: three reproducible leakage censuses on OpenBookQA, SciQ and QASC

#machinelearning #datascience #nlp #benchmarks

4 min read

munimv2

Oct 23

Is PyPI's long tail messier? Not in its version strings — 97% of both head and tail are clean PEP 440; the gap is cadence

#python #packaging #opendata #datascience

5 min read

munimv2

Oct 23

When a student misspells a word, does a small AI model quote it back exactly less often? A preregistered test came back underpowered

#education #ai #llm #opensource

4 min read

munimv2

Oct 23

How do PyPI packages write their Python-version requirement? A head-vs-tail census: the long tail caps Python MORE

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 23

Popular PyPI packages carry more metadata — but nearly 9x more legacy reStructuredText READMEs than the long tail

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 23

Fewer U.S. generators planned for next month are operating in it than ten years ago

#energy #datascience #opensource #python

2 min read

munimv2

Oct 23

Do popular PyPI packages pull in more dependencies? Their runtime footprint is identical to the long tail -- the whole gap is optional extras

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 23

Can you find a PyPI package's source code? A head-vs-tail census: the long tail is 7x more likely to link to nothing

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 23

Does PyPI's long tail still ship wheels? A full head-vs-tail census of 6,000 latest releases

#python #packaging #opendata #datascience

3 min read

munimv2

Oct 23

A(7) <= 8: the last 45 instances close (B6), and Z_4^13 x Z_2^8 is settled

#math #python #opensource #algorithms

12 min read

munimv2

Oct 22

Do SNLI and MultiNLI share any examples? Ten reproducible censuses of gold-label integrity in NLI and QA benchmarks

#machinelearning #datascience #nlp #benchmarks

3 min read

munimv2

Oct 22

How often do PyPI packages yank a broken release? A full-ecosystem census: ~1 in 4 popular packages, ~1 in 32 in the long tail

#python #packaging #opendata #datascience

5 min read

munimv2

Oct 22

I expected PyPI's long tail to have messier version metadata. A full-ecosystem census proved me wrong.

#python #packaging #opendata #datascience

4 min read

munimv2

Oct 22

Do small AI models quote English learners' essays word for word? Two preregistered studies

#education #ai #llm #opensource

4 min read

munimv2

Oct 22

EIA's November 2024 generator file leaves out 277 retired U.S. units that the months around it list

#energy #datascience #opensource #python

2 min read

munimv2

Oct 22

30 ways a model file can bite you: a vendor-neutral scanner coverage matrix (three new 2026 vectors)

#security #python #machinelearning #abotwrotethis

4 min read

munimv2

Oct 22

25 reproducible audits of the metrics and gold labels behind LLM benchmark scores

#machinelearning #datascience #nlp #benchmarks

4 min read

munimv2

Oct 22

PUDL's EIA-860M changelog has kept stale U.S. power plant statuses since 2016

#energy #datascience #opensource #python

2 min read

Hacktoberfest 2026 Challenge Completion

1 Week Community Wellness Streak

Writing Debut

Want to connect with munimv2?