Papers by Thomas Gueudre

5 papers
Sharing Encoder Representations across Languages, Domains and Tasks in Large-Scale Spoken Language Understanding (2023.acl-industry)

Copied to clipboard

Challenge: Larger encoders can improve accuracy for spoken language understanding (SLU) but are difficult to use given the inference latency constraints of online systems.
Approach: They propose to use a larger 170M parameter BERT encoder that shares representations across languages, domains and tasks for SLU.
Outcome: The proposed encoders achieve state-of-the-art performance on numerous NLP tasks.
Detecting and Mitigating Challenges in Zero-Shot Video Summarization with Video LLMs (2025.findings-acl)

Copied to clipboard

Challenge: Video Large Language Models (VLLMs) exhibit impressive zero-shot capabilities in video analysis, but their performance varies significantly depending on the LLM prompt, the characteristics of the video, and the properties of the training data and LLM architecture.
Approach: They propose to use Chain-of-Thought prompting to inject knowledge extracted by external, lightweight models into video summarization benchmarks to evaluate their performance.
Outcome: The proposed solutions improve summarization performance by injecting knowledge extracted by external, lightweight models.
Privacy Preserving Data Selection for Bias Mitigation in Speech Models (2025.acl-industry)

Copied to clipboard

Challenge: Existing methods for identifying subgroups raise privacy concerns and gather sensitive information at runtime might be impractical.
Approach: They propose a method to identify and train underperforming subgroups and train a model to predict if an utterance belongs to these subgroup.
Outcome: The proposed method reduces biases and improves performance on intent classification and automatic speech recognition tasks.
Knowledge Distillation Transfer Sets and their Impact on Downstream NLU Tasks (2022.emnlp-industry)

Copied to clipboard

Challenge: Domain Classification (DC) and Intent Classification/Named Entity Recognition (ICNER) are the most common methods for reducing teacher-student knowledge into manageable sizes for low-latency downstream applications.
Approach: They investigate whether distillation from a generic LM benefits downstream tasks . a domain classification and a task-specific data set are used to fine tune the model .
Outcome: The proposed model improves across tasks and test sets when only task-specific data is used.
Constraining word alignments with posterior regularization for label transfer (2022.naacl-industry)

Copied to clipboard

Challenge: Unsupervised word alignments are not always possible in industrial NLP pipelines, where multilingual annotation guidelines are complex and deviate from semantic consistency due to various factors.
Approach: They propose to constrain word alignment models to remain consistent with both source and target annotation guidelines by leveraging posterior regularization and labeled examples.
Outcome: The proposed model improves on the multiATIS++ dataset over AWESoME, and even a small amount of target language annotations can help.

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