Pavlo Kuchmiichuk

Pavlo Kuchmiichuk

I'm a PhD student in Computer Science at the University of Rochester, working in the Formal and Computational Semantics lab (FACTS.lab) advised by Aaron White.

My work bridges formal semantics and NLP: I explore how theories of meaning can enhance machine reasoning, and how computational methods can help us refine those theories.

Experience

CV

Download PDF →

Employment

Grammarly (now Superhuman)
Computational Linguist — 2018–2022

At Grammarly, I worked on the core product, implementing and shipping suggestions for improving grammar, style, clarity, and tone of the text. I also led a subteam of computational linguists, handling project planning, task prioritization, and technical supervision.

Education

University of Rochester
PhD student, Computer Science — 2024–present
University of Rochester
MS, Computational Linguistics — 2022–2024

Master's project: Toward Logical Form Induction: Encoding and Decoding Types.

Fulbright Graduate Student Program grantee.

Taras Shevchenko National University of Kyiv
MA, Applied Linguistics — 2020–2022

Thesis: Automatic error correction in Ukrainian text with the help of machine learning models.

Taras Shevchenko National University of Kyiv
BA, Applied Linguistics — 2016–2020

Thesis: Automatic detection and correction of spelling errors in Ukrainian text based on the noisy channel model.

Teaching

University of Rochester
Graduate Teaching Assistant
  • CSC186: Video Game Development (Spring 2025)
  • CSC252/452: Computer Organization (Fall 2025)
  • CSC186: Video Game Development (Spring 2026)

Research Interests

Computational Semantics

Logical Form Induction

I am broadly interested in how formal semantics and computational methods can inform and advance each other. Currently, I am working in FACTS.lab on the LoFI project, exploring how to integrate logical reasoning capabilities into AI systems.

Information Extraction

Event-Keyed Summarization
Gantt, Martin, Kuchmiichuk & White — EMNLP 2024 Findings

We introduce a novel task, event-keyed summarization (EKS), the goal of which is to produce a contextualized summary anchored to a specific event given a document and an extracted event structure.

TextData
Cross-Document Event-Keyed Summarization
Walden, Kuchmiichuk, Martin, Jin, Cao, Sun, Allen & White — XLLM 2025

We extend event-keyed summarization to the cross-document setting, in which summaries must synthesize information from accounts of the same event as given by multiple sources.

TextData

Ukrainian NLP

Silver Data for Coreference Resolution in Ukrainian: Translation, Alignment, and Projection
Kuchmiichuk — UNLP 2023

I present two silver datasets for coreference resolution in Ukrainian, adapted from existing English data by manual translation and machine translation in combination with automatic alignment and annotation projection.

TextData
UA-GEC: Grammatical Error Correction and Fluency Corpus for the Ukrainian Language
Syvokon, Nahorna, Kuchmiichuk & Osidach — UNLP 2023

We have built a large-scale corpus for grammatical error correction and fluency in Ukrainian. UA-GEC includes texts from a diverse pool of contributors, annotated by professional proofreaders for errors relating to fluency, grammar, punctuation, and spelling.

TextData

Grammatical Error Correction

UA-GEC: Grammatical Error Correction and Fluency Corpus for the Ukrainian Language
Syvokon, Nahorna, Kuchmiichuk & Osidach — UNLP 2023

We have built a large-scale corpus for grammatical error correction and fluency in Ukrainian. UA-GEC includes texts from a diverse pool of contributors, annotated by professional proofreaders for errors relating to fluency, grammar, punctuation, and spelling.

TextData

Contact

  • Email
  • Google Scholar
  • LinkedIn
  • ORCID
  • Github