Chalmers University of Technology

2 related articles

arXiv Oct 9, 2026
D-SLR matches or beats truncated SVD at three SVDs and cuts prediction changes on LLM embedding tables by about tenfold
The authors propose the Disjoint Row-Sparse plus Low-Rank (D-SLR) decomposition, in which each row is either stored verbatim or fitted by a shared low-rank basis but never both; under squared error this restriction is lossless, the algorithm scores every rank and stored-row count in closed form at about three SVDs, and it provides an assumption-free a-posteriori lower-bound certificate; on synthetic matrices and LLM embedding tables D-SLR is never worse than the truncated SVD at equal cost and, at half the table's parameters, changed about ten times fewer model predictions.
The authors propose the Disjoint Row-Sparse plus Low-Rank (D-SLR) decomposition, in which each row is either stored verbatim or fitted by a shared low-rank basis but never both; under squared error this restriction is lossless, the algorithm scores every rank and stored-row count in closed form at about three SVDs, and it provides an assumption-free a-posteriori lower-bound certificate; on synthetic matrices and LLM embedding tables D-SLR is never worse than the truncated SVD at equal cost and, at half the table's parameters, changed about ten times fewer model predictions.
Chalmers University of Technology
Large Language Model Low-Rank Matrix Factorization Sparse plus Low-Rank Decomposition
bioRxiv Sep 29, 2026
TRIDENT-2 predicts chemical toxicity across Eukaryota from 560,780 assays with average median absolute error of 1.76 to 3.80
The authors present TRIDENT-2, a multimodal artificial intelligence model trained on 560,780 toxicity assays spanning 82,775 chemicals, 6,793 species, and multiple exposure scenarios to predict chemical toxicity across evolutionarily diverse eukaryotic species, reporting an average median absolute error of 1.76 to 3.80 and remaining accurate across broad chemical and taxonomic distances, which allows toxicity assessment for species and chemicals beyond current experimental evidence.
The authors present TRIDENT-2, a multimodal artificial intelligence model trained on 560,780 toxicity assays spanning 82,775 chemicals, 6,793 species, and multiple exposure scenarios to predict chemical toxicity across evolutionarily diverse eukaryotic species, reporting an average median absolute error of 1.76 to 3.80 and remaining accurate across broad chemical and taxonomic distances, which allows toxicity assessment for species and chemicals beyond current experimental evidence.
University of Gothenburg Chalmers University of Technology Science for Life Laboratory RWTH Aachen University Mercedes Dalman Patrik Svedberg Thomas Backhaus Jens Henriksson Erik Kristiansson Styrbjörn Käll Mikael B. Gustavsson

You’re all caught up

D-SLR matches or beats truncated SVD at three SVDs and cuts prediction changes on LLM embedding tables by about tenfold

TRIDENT-2 predicts chemical toxicity across Eukaryota from 560,780 assays with average median absolute error of 1.76 to 3.80