Andreas Madsen

14 papers A* 2Journal 9Unranked 3
YearRankTypeTitle / Venue / Authors
2024 — conf
Are self-explanations from Large Language Models faithful?
ACL (Findings)
Andreas Madsen, Sarath Chandar, Siva Reddy
2024 J jnl
Are self-explanations from Large Language Models faithful?
CoRR
Andreas Madsen, Sarath Chandar, Siva Reddy
2024 — conf
Data Interface for an Interactable Ship Bridge towards Maritime Autonomous Surface Ships at Human-in-the-loop Levels.
ICHMS
Baiheng Wu, Lars Ivar Hatledal, Andreas Madsen, Tongtong Wang, Andreas Brandsæter, Ottar L. Osen
2024 A* conf
Faithfulness Measurable Masked Language Models.
ICML
Andreas Madsen, Siva Reddy, Sarath Chandar
2024 J jnl
Interpretability Needs a New Paradigm.
CoRR
Andreas Madsen, Himabindu Lakkaraju, Siva Reddy, Sarath Chandar
2024 J jnl
New Faithfulness-Centric Interpretability Paradigms for Natural Language Processing.
CoRR
Andreas Madsen
2023 J jnl
Faithfulness Measurable Masked Language Models.
CoRR
Andreas Madsen, Siva Reddy, Sarath Chandar
2023 J jnl
Post-hoc Interpretability for Neural NLP: A Survey.
ACM Comput. Surv.
Andreas Madsen, Siva Reddy, Sarath Chandar
2022 — conf
Evaluating the Faithfulness of Importance Measures in NLP by Recursively Masking Allegedly Important Tokens and Retraining.
EMNLP (Findings)
Andreas Madsen, Nicholas Meade, Vaibhav Adlakha, Siva Reddy
2021 J jnl
Evaluating the Faithfulness of Importance Measures in NLP by Recursively Masking Allegedly Important Tokens and Retraining.
CoRR
Andreas Madsen, Nicholas Meade, Vaibhav Adlakha, Siva Reddy
2021 J jnl
Post-hoc Interpretability for Neural NLP: A Survey.
CoRR
Andreas Madsen, Siva Reddy, Sarath Chandar
2020 A* conf
Neural Arithmetic Units.
ICLR
Andreas Madsen, Alexander Rosenberg Johansen
2020 J jnl
Neural Arithmetic Units.
CoRR
Andreas Madsen, Alexander Rosenberg Johansen
2019 J jnl
Measuring Arithmetic Extrapolation Performance.
CoRR
Andreas Madsen, Alexander Rosenberg Johansen