Papers by Vésteinn Snæbjarnarson

6 papers
Activation Scaling for Steering and Interpreting Language Models (2024.findings-emnlp)

Copied to clipboard

Challenge: a successful intervention should flip the correct with the wrong token, while remaining sparse.
Approach: They propose to use activation scaling to flip the correct with the wrong token . they use gradient-based optimization to learn and evaluate a specific kind of efficient intervention .
Outcome: The proposed method performs comparable with steering vectors but is much less minimal.
Context versus Prior Knowledge in Language Models (2024.acl-long)

Copied to clipboard

Challenge: Existing studies have investigated how often a model will rely on prior knowledge over conflicting contextual information in answering questions.
Approach: They propose two mutual information-based metrics to measure a model’s dependency on a context and on its prior about an entity.
Outcome: The proposed metrics show that language models can integrate prior knowledge and new information in a predictable way across different questions and contexts.
A Warm Start and a Clean Crawled Corpus - A Recipe for Good Language Models (2022.lrec-1)

Copied to clipboard

Challenge: Pre-trained neural language models have shown impressive results when adapted for a variety of classification and text generation tasks.
Approach: They propose to use Icelandic's Icelandic Common Crawl Corpus to train language models that achieve state-of-the-art performance in downstream tasks.
Outcome: The proposed models achieve state-of-the-art in a variety of downstream tasks including part-of speech tagging, named entity recognition and constituency parsing.
Byte-Level Grammatical Error Correction Using Synthetic and Curated Corpora (2023.acl-long)

Copied to clipboard

Challenge: Spelling mistakes due to typos and rushed writing, nonstandard punctuation and spelling, and grammatical and stylistic issues are common to almost everyone who writes any kind of text.
Approach: They propose to use a common subword unit vocabulary and byte-level encoding to fine tune two subword-level models and one byte level model on hand-corrected error corpora.
Outcome: The proposed model improves accuracy for spelling and grammatical errors and more complex errors.
On the Proper Treatment of Units in Surprisal Theory (2026.acl-long)

Copied to clipboard

Challenge: empirical work often leaves the notion of a unit underspecified . empirical work has sought to characterize the processing difficulty comprehenders experience .
Approach: They propose a framework for reasoning about surprisal over arbitrary unit inventories . they argue that surprises should be explicit and treat tokenization as implementation detail .
Outcome: The proposed framework disentangles the models' definitions and the regions of interest and treats tokenization as an implementation detail rather than a scientific primitive.
Natural Questions in Icelandic (2022.lrec-1)

Copied to clipboard

Challenge: Developing such datasets is important for the development and evaluation of Icelandic QA systems.
Approach: They present the first extractive question answering dataset for Icelandic, Natural Questions in Icelandic.
Outcome: The proposed dataset is a valuable resource for Icelandic which is being evaluated by a team of researchers.

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