logo-lab/trl-bench-ckpts

TRL-Bench pretrained checkpoints

Mirror of upstream pretrained weights for models evaluated in TRL-Bench (paper: arXiv:2606.09323; code: LOGO-CUHKSZ/TRL-Bench). Each subdirectory contains:

• a copy of the upstream LICENSE
• a NOTICE file with citation, provenance, and SHA256
• the binary checkpoint(s)

The checkpoints have not been modified — they are byte-identical to the upstream releases. This repo exists only to provide a reliable, scriptable download path that does not depend on third-party file-share services (Google Drive, SharePoint) which can rate-limit or expire links.

Coverage

Model	License (upstream)	Citation	Note
TUTA	MIT (Microsoft)	Wang et al., KDD 2021
TURL	Apache-2.0 (sunlab-osu)	Deng et al., PVLDB 14(3), 2020
TaBERT	CC BY-NC 4.0 (Facebook AI Research)	Yin et al., ACL 2020	Non-commercial only — see tabert/NOTICE

TaBERT is mirrored under the upstream CC BY-NC 4.0 terms strictly for non-commercial reproduction of the paper experiments. If you intend a commercial application you may not use that checkpoint; train your own or seek an alternate license from the original authors (see tabert/NOTICE).

Other TRL-Bench upstream models — TabSketchFM (CC BY-NC-ND 4.0, upstream only), Starmie (user-trained), and TABBIE (obtain from upstream) — are not mirrored here. See docs/CHECKPOINT_LICENSES.md and scripts/download_checkpoints.sh in the TRL-Bench repo for the upstream paths used for those models.

Downloading

Programmatic (huggingface_hub):

from huggingface_hub import snapshot_download
snapshot_download(
    "logo-lab/trl-bench-ckpts",
    allow_patterns=["tuta/*", "turl/*", "tabert/*"],
    local_dir="./checkpoints",
)

Or via TRL-Bench's bundled script (verifies SHA256 against scripts/checksums.sha256 after download):

git clone https://github.com/LOGO-CUHKSZ/TRL-Bench.git
cd TRL-Bench
bash scripts/download_checkpoints.sh

Layout

tuta/
  LICENSE              # MIT (upstream Microsoft)
  NOTICE               # citation, provenance, SHA256
  tuta.bin             # 511 MB
turl/
  LICENSE              # Apache-2.0 (upstream sunlab-osu)
  NOTICE               # citation, provenance, SHA256
  pretrained/
    config.json
    pytorch_model.bin  # 1.2 GB
tabert/
  LICENSE              # CC BY-NC 4.0 (upstream Facebook AI Research)
  NOTICE               # citation, provenance, SHA256 + NC terms
  tabert_base_k3/
    model.bin          # 266 MB
    tb_config.json
    version.txt

These paths match the expectations of TRL-Bench's scripts/checksums.sha256 and the per-model wrappers in src/trl_bench/models/{tuta,turl,tabert}/.

Attribution

This is a third-party mirror. The upstream authors are the sole copyright holders of the model weights. The per-model LICENSE and NOTICE files in each subdirectory are authoritative.

If you use these checkpoints, cite the upstream papers (see each subdir's NOTICE). If you use TRL-Bench, also cite:

@article{pang2026trl,
  title={TRL-Bench: Standardizing Cross-Paradigm Representation-Level Evaluation of Tabular Encoders},
  author={Pang, Wei and Jian, Xiangru and Li, Hehan and Yu, Zhixuan and Xue, Alex and Li, Jinyang and Dong, Zhengyuan and Zhao, Xinjian and Xu, Hao and Zhang, Chao and Cheng, Reynold and {\"O}zsu, M. Tamer and Yu, Tianshu},
  journal={arXiv preprint arXiv:2606.09323},
  year={2026}
}