R Observatory: Metrics for FMAT version 2026.1

Balamuta, James Joseph

doi:10.5281/zenodo.21843040

FMAT

2026.1Package record

The Fill-Mask Association Test

0packages depend

5.6Kdownloads / year

–test coverage

13/13flavors OK

Overview

About

Maintained by Han Wu Shuang BaoCRAN page ↗GitHub ↗Website ↗Issues ↗

The Fill-Mask Association Test ('FMAT') doi:10.1037/pspa0000396 is an integrative, probability-based social computing method using Masked Language Models to measure conceptual associations (e.g., attitudes, biases, stereotypes, social norms, cultural values) as propositional semantic representations in natural language. Supported language models include 'BERT' doi:10.48550/arXiv.1810.04805 and its variants available at 'Hugging Face' https://huggingface.co/models?pipeline_tag=fill-mask. Methodological references and installation guidance are provided at https://psychbruce.github.io/FMAT/.

Changes in 2026.1

This version brings crucial changes and improvements to the add.tokens method of FMAT_run(). Old versions are completely deprecated.

Deprecated the add.method parameter of FMAT_run(): Now it always computes **average** subword token embeddings, with another new parameter weight.decay (default value = 1, i.e., equally weighted) that can specify the relative importance of multiple subwords.
A smaller decay value would give greater weight to the former subwords than to the latter subwords.
The *i*-th subword $w_i$ has the raw weight score as $\text{weight}(w_i) = decay^i$.
See also weight_decay() for computational details.
Added special_case() as an explicit function to better specify models requiring special treatment, which was previously just a parameter of FMAT_run().
Improved the internal technical treatment for subword token combination when "\\u0120" is the special prefix. This is important for more accurate out-of-vocabulary fill-mask probability estimates of whole words and phrases, mainly relevant to models using the RoBERTa architecture and only influencing results under add.tokens=TRUE. Note that FMAT_run() results with add.tokens=FALSE are not affected by this change.
Improved progress output information of FMAT_run().
Improved ICC_models() to support ICC estimates of both log probability (raw) and log probability ratio (LPR).

From NEWS.md in the release

Install

R (CRAN)pak (dev)Ubuntu (r2u)Fedora (COPR)openSUSE (OBS)

install.packages("FMAT", repos = "https://cloud.r-project.org/")

Installs the current CRAN release. Prebuilt binaries on Windows and macOS.

if (!requireNamespace("pak", quietly = TRUE))
  install.packages("pak", repos = "https://cloud.r-project.org/")
pak::pak("psychbruce/FMAT")

Development version from psychbruce/FMAT. Installs pak first if it is not already available.

Release

# enable cranapt (r2u) if it is not already set up, then install
grep -rqs r2u.stat.illinois.edu /etc/apt/sources.list /etc/apt/sources.list.d/ \
  || curl -fsSL https://raw.githubusercontent.com/eddelbuettel/r2u/master/inst/scripts/add_cranapt_noble.sh | sudo bash
sudo apt install -y r-cran-fmat

Fast Ubuntu binaries via r2u (no compilation). The cranapt setup runs once and is skipped when it is already enabled. We list this channel from recorded r2u download activity rather than from a live index of the repository, so treat it as last known rather than guaranteed.

bspm (used by r2u) also needs make: sudo apt install make

# enable the iucar/cran COPR if it is not already present, then install
ls /etc/yum.repos.d/*iucar*cran*.repo >/dev/null 2>&1 \
  || sudo dnf copr enable -y iucar/cran
sudo dnf install -y R-CRAN-FMAT

RPMs from the iucar/cran COPR, rebuilt from CRAN.

For install.packages() integration in R, also: sudo dnf install R-CoprManager

# add the autoCRAN OBS repo (Tumbleweed) if it is missing, then install
zypper -q lr | grep -qi autoCRAN \
  || sudo zypper addrepo https://download.opensuse.org/repositories/devel:/languages:/R:/autoCRAN/openSUSE_Tumbleweed/devel:languages:R:autoCRAN.repo
sudo zypper install -y R-FMAT

RPMs auto-built from CRAN on openSUSE OBS (Tumbleweed). We list this channel from recorded autoCRAN download activity rather than from a live index of the repository, so treat it as last known rather than guaranteed.

Health

CRAN checks

13OK

CRAN check page ↗

Slowest check: 1.5 min · r-oldrel-windows-x86_64

Status changes

OK13 OK2026-08-05
NOTE12 OK, 1 NOTE2026-08-01
OK13 OK2026-06-09
ERROR12 OK, 1 ERROR2026-06-08
OK14 OK2026-03-10

Advisories

none known

How this is measured·How advisories are found

Documentation

In release 2026.1

READMEYes · 1,330 wordsVignettesNoNEWSYes · 100% structured · mentions 2026.1Citation fileNoWebsiteYes · pkgdown

Help pages: 17 (2 internal)includes a package overview page
Examples that run: 23%3 run in a plain check · 2 only when donttest examples are enabled (R CMD check --as-cran) · 7 never run · 1 hold no code
Exported functions with examples: 80%
Documented parameters: 87%
Return-value docs: 93%
Help pages citing references: 0%

How these are measured

Downloads

5.6K

CRAN downloads in the past year

Rank #8,284 · ~15/day · ~468/mo

Releases R versions

Daily download trend is not available in this view yet.

47830 days

1.1K90 days

5.6K1 year

Compare downloads with other packages →

Also on89 r2u6 autocran

How downloads are counted

Repository

psychbruce/fmat ↗Account: psychbruce

16Stars

1Forks

0Open issues

0Open PRs

0Releases

63Commits

1Commit identities

bertbert-modelbert-modelscontextualized-representationlarge-language-modelsmasked-language-modelsnatural-language-processingnatural-language-understanding

License file reads as GPL-3.0 · 63 commits · Last activity 2026-03-07 · 0% stars, 30d

Responsiveness

How these are sampled

PRs merged (all time): not measured
Median days to close, 50 most recently updated closed issues: 103
Median days to close, 50 most recently updated closed PRs: not measured

Stars over time

2025-08-02 · 152026-07-07 · 16

How these are measured

PracticesBeta

8 development-tooling and community-health practices detected across 4 families in the upstream repository

Checks run against github.com/psychbruce/fmat on 2026-10-18.

Continuous integration (2)

GitHub ActionsR CMD check workflow

The R CMD check workflow runs on Linux

Docs source (3)

pkgdown

GitHub Pages site last deployed 2026-03-07 · The repository has a site deploy workflow

pkgdown site last built 2026-03-07

CRAN release process (2)

cran-comments.mdCRAN-SUBMISSION

Lint, format, editor (1)

RStudio project

How this is detected·Detection ruleset v3 (2026-09-26)

Dependencies

Declared dependencies

18 external dependencies (excludes base and recommended)

Installing FMAT adds 46 packages (28 with compiled code, 34 MB of source tarballs). Heaviest: stringi 11 MB, data.table 5.7 MB, Rcpp 1.8 MB.

Depends (1)

R >= 4.0.0

Imports (14)

reticulate data.table stringr forcats rvest psych irr glue crayon cli purrr plyr dplyr tidyr

LinkingTo (0)

none

Suggests (5)

bruceR PsychWordVec text sweaternlme

Enhances (0)

none

Dependency license breakdown →

Reverse dependencies

0direct

0indirect

Nothing depends on this yet.

How these are counted

Code & Tests

People & History

People (1)

Han Wu Shuang Bao

Author, Maintainer · added in 2025.8

Maintainer since 2023.8 (2023)

Listed in earlier versions (1)

Han-Wu-Shuang Bao

no longer listed · 2023.8 to 2025.4

Releases (11)

Pick two releases to compare their code metrics.

R
R 4.6.0 released · 2026-04-24
2026.1Latest
record
2026-01-12 · current release · diff ↗ ·
added 2 exports
special_case, weight_decay
2025.12
record
2025-12-15 · diff ↗
2025.8
record
2025-08-19 · diff ↗
R
R 4.5.0 released · 2025-04-11
2025.4
record
2025-04-08 · diff ↗ ·
removed 1 export
FMAT_load

added 3 exports
BERT_remove, fill_mask, fill_mask_check
2025.3
record
2025-03-19 · diff ↗ ·
added 2 exports
BERT_info_date, set_cache_folder
2024.7
record
2024-07-29 · diff ↗
2024.6
record
2024-06-12 · diff ↗ ·
removed 1 export
cc
2024.5
record
2024-05-19 · diff ↗ ·
added 1 export
BERT_info
2024.4
record
2024-04-30 · diff ↗ ·
added 2 exports
BERT_vocab, ICC_models
R
R 4.4.0 released · 2024-04-24
2024.3
record
2024-03-22 · diff ↗ ·
added 1 export
BERT_download
2023.8
record
2023-08-11
R
R 4.3.0 released · 2023-04-21

Package metadata

First published: 2023-08-11
Total releases: 11 / 3 yrs
License: GPL-3 OSI
Minimum R: ≥ 4.0.0
Download size: 34 KB

Cite

Cite this package

Run in R for the authors' preferred citation:

citation("FMAT")

Format

Bao, H. W. S. (2026). FMAT: The Fill-Mask Association Test (Version 2026.1) [Computer software]. https://doi.org/10.32614/CRAN.package.FMAT

This package has no citation file, so citation() returns this entry built from DESCRIPTION.

Cite the R Observatory

For a number measured here: a download total, a coverage figure, an archival date.

Format

APA

Balamuta, J. J. (2026). R Observatory: Metrics for FMAT version 2026.1 [Data set]. HJJB, LLC. Data release v2026-10-22. https://doi.org/10.5281/zenodo.21843040

From data release v2026-10-22, which the citation names so these numbers can be found later. More on citing and the projects behind them.

Report a problem with this page →