Papers by Petr Motlicek

8 papers
Claim-Dissector: An Interpretable Fact-Checking System with Joint Re-ranking and Veracity Prediction (2023.findings-acl)

Copied to clipboard

Challenge: a novel latent variable model for fact-checking and analysis learns to identify the veracity of a claim and its relevant evidences.
Approach: They propose to disentangle the per-evidence relevance probability and its contribution to the final veracity probability in an interpretable way.
Outcome: The proposed model can achieve competitive results on the FEVER dataset while using significantly fewer parameters.
Dialog2Flow: Pre-training Soft-Contrastive Action-Driven Sentence Embeddings for Automatic Dialog Flow Extraction (2024.emnlp-main)

Copied to clipboard

Challenge: Dialog2Flow embeddings allow for modeling dialogs as continuous trajectories in a latent space with distinct action-related regions.
Approach: They propose dialog2Flow embeddings that map dialogs to a latent space and cluster them according to their communicative and informative functions.
Outcome: The proposed workflow embeddings show superior performance across domains.
SDialog: A Python Toolkit for End-to-End Agent Building, User Simulation, Dialog Generation, and Evaluation (2026.eacl-demo)

Copied to clipboard

Challenge: SDialog is an open-source Python toolkit for end-to-end development, simulation, evaluation and analysis of LLM-based conversational agents.
Approach: They present an open-source Python toolkit for end-to-end development, simulation, evaluation and analysis of LLM-based conversational agents.
Outcome: SDialog enables more controlled, transparent, and systematic research on conversational systems.
Abstract Text Summarization: A Low Resource Challenge (D19-1)

Copied to clipboard

Challenge: Existing datasets for multilingual text summarization are difficult to construct and lack of human knowledge and language processing abilities in computers makes text summaries a challenging task.
Approach: They propose an iterative data augmentation approach which uses synthetic data along with the real summarization data for the German language.
Outcome: The proposed system improves on the development and test sets on the German language text using the state-of-the-art “Transformer” model.
Idiap NMT System for WAT 2019 Multimodal Translation Task (D19-52)

Copied to clipboard

Challenge: In the past few decades, multi-modality has received critical attention in translation studies, although the benefit of visual modality in machine translation is still in debate.
Approach: They propose to use the Transformer model and IITB English-Hindi parallel corpus as additional data sources for the evaluation and challenge test sets.
Outcome: The proposed system outperforms systems that consider visual information in the English-Hindi Multi-Modal Translation task.
TokenVerse: Towards Unifying Speech and NLP Tasks via Transducer-based ASR (2024.emnlp-main)

Copied to clipboard

Challenge: Existing approaches to automatic speech recognition use cascaded pipelines for tasks like voice activity detection, diarization, transcription and subsequent processing.
Approach: They propose a single Transducer-based model that integrates task-specific tokens into the reference text during ASR model training, streamlining inference and eliminating the need for separate NLP models.
Outcome: The proposed model outperforms the existing pipeline on speaker change detection, endpointing, and NER tasks while outperforming the existing model in individual task performance.
Reliability Estimation of News Media Sources: Birds of a Feather Flock Together (2024.naacl-long)

Copied to clipboard

Challenge: Recent research has shown that predicting sources’ reliability is an important first-prior step in addressing additional challenges such as fake news detection and fact-checking.
Approach: They propose a method that leverages reinforcement learning strategies to estimate the reliability degree of news sources based on how all the news media sources interact with each other on the Web.
Outcome: The proposed method can predict reliability labels on a large news media reliability dataset.
Fast Streaming Transducer ASR Prototyping via Knowledge Distillation with Whisper (2024.findings-emnlp)

Copied to clipboard

Challenge: a recent study shows that training of ASR models with little to no supervised data is challenging.
Approach: They propose a framework to train streaming Transformer-Transducer models with pseudo-labeled (PL) speech from foundational speech models.
Outcome: The proposed framework can be trained from scratch with pseudo-labeled speech from foundational speech models (FSMs) the proposed framework is validated on 6 languages from CommonVoice and proposes multiple heuristics to filter out hallucinated PLs.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations