Workshop paper
ALLMTS: Automated LLM-based Medical Triage System
95% triage accuracy, against 47% for a plain LLM.
Publications
Papers by people on our team.
Research co-authored by people on the Vstorm team. One paper comes from a Vstorm client project. The others come from research at universities, in open-source projects and in earlier roles, and every co-author is listed on the card.
-
- Conference paper
Unifying Perspectives: Plausible Counterfactual Explanations on Global, Group-wise, and Local Levels
Local, group-wise and global counterfactuals.
- Preprint
CEL: Comprehensive Counterfactual Explanations Library and Benchmark
A library and benchmark: 18 datasets, 14 methods.
- Competition entry
Coding agents for automated metaheuristic design
Coding agents that design optimization algorithms.
- Conference paper
Plausible Visual Counterfactual Explanations in Latent Space with Normalizing Flows
Image counterfactuals via normalizing flows.
- Workshop paper
From Supra to Sub: A Hierarchical Margin-Based Approach for Audio Deepfake Source Tracing
Traces a fake voice to the model that made it.
- Preprint
Bielik-Minitron-7B: Compressing Large Language Models via Structured Pruning and Knowledge Distillation for the Polish Language
A Polish LLM cut from 11B to 7.35B parameters.
- Preprint
Automatic Design of Optimization Test Problems with Large Language Models
LLM-driven search for optimization test functions.
- Conference paper
Large Language Models for Data Extraction in Slot-Filling Tasks
Open-source LLMs compared on slot filling.
Work with us
Talk to our engineers
A 30-minute call with an engineer who assesses whether an agent fits your workflow.