TagoMind · Kamil Tagowski, PhD

About

Wed, 28 Oct 2026 00:00:00 +0000

I work at the seam between research and engineering. My academic background is in graph representation learning and natural language processing, and my day-to-day is building systems that take those ideas out of papers and into tools people rely on.

That has meant co-authoring LEPISZCZE, a NeurIPS 2022 benchmark that standardized how Polish language models are evaluated, and it has meant shipping production legal and tax AI platforms that index millions of documents and serve working researchers. I care about the unglamorous parts that decide whether a model is actually useful: clean datasets, honest evaluation, and software that does not fall over.

Work with me

Wed, 28 Oct 2026 00:00:00 +0000

Engagements range from a few days of advisory work to multi-month builds. I work independently and alongside in-house teams, owning a problem outright or plugging into an existing stack.

Schematize

Thu, 01 Oct 2026 00:00:00 +0000

Abstract

Empirical legal research often relies on turning research questions into structured data extracted from large collections of rulings and judgments. Designing the extraction schema and then extracting the data remain a manual, expertise-heavy bottleneck. We present schematize, an open-source multi-agent system that interactively turns a researcher’s problem statement into a validated extraction schema that can later be used for autonomous extraction. Schematize couples (i) a clarification dialogue that elicits implicit expert intent, (ii) iterative schema generation, (iii) data-grounded refinement that tests the schema against documents, and (iv) chat-based post-editing. We evaluated the system with human legal professionals, introducing our novel methodology, and schematize achieves top performance in most of tested configurations. While the system is designed to be domain-agnostic and applicable to any document collection, we tailor and evaluate it on legal research problems. We release schematize as a pip-installable Python package with full documentation.

FSDBench

Wed, 01 Jul 2026 00:00:00 +0000

Abstract

Before a tax authority can issue a ruling, it must receive a complete description of the taxpayer’s situation, yet no benchmark measures whether language models can systematically elicit all relevant facts through dialogue. We introduce FSDBench (Factual State Discovery Benchmark), in which a discovery agent questions a simulated taxpayer grounded in a real tax document. The dataset comprises 500 narratives from official Polish tax interpretations, decomposed into 32,874 atomic facts with validated supported precision (97.6%), atomicity (93.8%), and sentence coverage (96.0%). Experiments with four models show that even the best system recovers only 77% of facts on easy samples and under 49% on hard samples after 50 turns. These findings establish conversational fact elicitation as a challenging open problem requiring retrieval-augmented and adaptive questioning strategies.

Finance AI

Mon, 01 Jun 2026 00:00:00 +0000

An applied finance-AI research group in active development, building out the team and the system.

TACTYC

Mon, 01 Jun 2026 00:00:00 +0000

Toss A Coin to A Cartographer. AI that generates and edits game maps: vision-language models read the terrain tile by tile, generative models redraw it.

Bridging AI and Law

Fri, 01 May 2026 00:00:00 +0000

In brief

The paper presents a production-scale, multi-agent platform that bridges AI research and legal practice. It indexes more than 3 million legal and tax documents and over 300 million semantic vectors, and it exposes them to a Quantitative Legal Agent architecture that runs aggregation queries and produces interpretable analyses across jurisdictions, rather than answering one question about one document at a time.

Production legal-analytics platforms of this kind combine large-scale document retrieval, expert-defined criteria, and structured extraction to support quantitative studies. That combination is the foundation that later work such as Schematize builds on when it automates the schema-design step.

Flat Explorer

Thu, 01 Jan 2026 00:00:00 +0000

AI-assisted flat hunting: scores apartment listings, summarizes visits, and compares options side by side.

Unite! Researcher Discovery

Thu, 01 Jan 2026 00:00:00 +0000

Semantic search and visualization platform for 200k+ researcher profiles across 9 European universities.

AI-Tax

Wed, 01 Jan 2025 00:00:00 +0000

A Polish tax & legal AI platform built at Wrocław Tech: AI search, chat with citations, and agentic extraction over tax interpretations and court judgments.

JuDDGES

Mon, 01 Jan 2024 00:00:00 +0000

EU-funded platform for NLP analysis of judicial decisions across Poland and England & Wales.

RAFEN

Sat, 01 Jul 2023 00:00:00 +0000

Abstract

Learning representations of nodes has been a crucial area of the graph machine learning research area. A well-defined node embedding model should reflect both node features and the graph structure in the final embedding. In the case of dynamic graphs, this problem becomes even more complex as both features and structure may change over time. The embeddings of particular nodes should remain comparable during the evolution of the graph, what can be achieved by applying an alignment procedure. This step was often applied in existing works after the node embedding was already computed. In this paper, we introduce a framework, RAFEN, that allows to enrich any existing node embedding method using the aforementioned alignment term and learning aligned node embedding during training time. We propose several variants of our framework and demonstrate its performance on six real-world datasets. RAFEN achieves on-par or better performance than existing approaches without requiring additional processing steps.

LEPISZCZE

Mon, 28 Nov 2022 00:00:00 +0000

Abstract

The availability of compute and data to train larger and larger language models increases the demand for robust methods of benchmarking the true progress of LM training. Recent years witnessed significant progress in standardized benchmarking for English. Benchmarks such as GLUE, SuperGLUE, or KILT have become de facto standard tools to compare large language models. Following the trend to replicate GLUE for other languages, the KLEJ benchmark has been released for Polish. In this paper, we evaluate the progress in benchmarking for low-resourced languages. We note that only a handful of languages have such comprehensive benchmarks. We also note the gap in the number of tasks being evaluated by benchmarks for resource-rich English/Chinese and the rest of the world. In this paper, we introduce LEPISZCZE (the Polish word for glew, the Middle English predecessor of glue), a new, comprehensive benchmark for Polish NLP with a large variety of tasks and high-quality operationalization of the benchmark. We design LEPISZCZE with flexibility in mind. Including new models, datasets, and tasks is as simple as possible while still offering data versioning and model tracking. In the first run of the benchmark, we test 13 experiments (task and dataset pairs) based on the five most recent LMs for Polish. We use five datasets from the Polish benchmark and add eight novel datasets. As the paper’s main contribution, apart from LEPISZCZE, we provide insights and experiences learned while creating the benchmark for Polish as the blueprint to design similar benchmarks for other low-resourced languages.

CLARIN-PL Embeddings

Sat, 01 Jan 2022 00:00:00 +0000

Transformer-based NLP library for text classification, sequence labeling, and embeddings in Polish.

FILDNE

Sat, 01 Jan 2022 00:00:00 +0000

Abstract

Representation learning on graphs has emerged as a powerful mechanism to automate feature vector generation for downstream machine learning tasks. The advances in representation on graphs have centered on both homogeneous and heterogeneous graphs, where the latter presenting the challenges associated with multi-typed nodes and/or edges. In this paper, we consider the additional challenge of evolving graphs. We ask the question of whether the advances in representation learning for static graphs can be leveraged for dynamic graphs and how? It is important to be able to incorporate those advances to maximize the utility and generalization of methods. To that end, we propose the Framework for Incremental Learning of Dynamic Networks Embedding (FILDNE), which can utilize any existing static representation learning method for learning node embeddings while keeping the computational costs low. FILDNE integrates the feature vectors computed using the standard methods over different timesteps into a single representation by developing a convex combination function and alignment mechanism. Experimental results on several downstream tasks, over seven real-world data sets, show that FILDNE is able to reduce memory (up to 6x) and computational time (up to 50x) costs while providing competitive quality measure gains (e.g., improvements up to 19 pp AUC on link prediction and up to 33 pp mAP on graph reconstruction) with respect to the contemporary methods for representation learning on dynamic graphs.

LEPISZCZE

Sat, 01 Jan 2022 00:00:00 +0000

A 14-task NLP benchmark for Polish with unified evaluation infrastructure and public leaderboard.

Embedding Alignment Methods in Dynamic Networks

Wed, 16 Jun 2021 00:00:00 +0000

Abstract

In recent years, dynamic graph embedding has attracted a lot of attention due to its usefulness in real-world scenarios. In this paper, we consider discrete-time dynamic graph representation learning, where embeddings are computed for each time window, and then are aggregated to represent the dynamics of a graph. However, independently computed embeddings in consecutive windows suffer from the stochastic nature of representation learning algorithms and are algebraically incomparable. We underline the need for embedding alignment process and provide nine alignment techniques evaluated on real-world datasets in link prediction and graph reconstruction tasks. Our experiments show that alignment of Node2vec embeddings improves the performance of downstream tasks up to 11 pp compared to the not aligned scenario.

Incremental Learning in Dynamic Networks for Node Classification

Mon, 11 Sep 2017 00:00:00 +0000

In brief

The paper proposes an incremental, active-scheme method for multi-class classification of node states that change over time and depend on information spreading through the network. It is demonstrated on a social network dataset in which messages exchanged between users update the feature vector of each receiver, and the task is to predict the receiver’s next state. The authors frame the setting as incremental learning in a non-stationary network environment and report that the approach forecasts future node states effectively on real data.