Search Results for author: Debjit Paul

Found 13 papers, 10 papers with code

REFINER: Reasoning Feedback on Intermediate Representations

1 code implementation4 Apr 2023 Debjit Paul, Mete Ismayilzada, Maxime Peyrard, Beatriz Borges, Antoine Bosselut, Robert West, Boi Faltings

Language models (LMs) have recently shown remarkable performance on reasoning tasks by explicitly generating intermediate inferences, e. g., chain-of-thought prompting.

Ranking and Selecting Multi-Hop Knowledge Paths to Better Predict Human Needs

1 code implementation NAACL 2019 Debjit Paul, Anette Frank

To make machines better understand sentiments, research needs to move from polarity identification to understanding the reasons that underlie the expression of sentiment.

Common Sense Reasoning

COINS: Dynamically Generating COntextualized Inference Rules for Narrative Story Completion

1 code implementation ACL 2021 Debjit Paul, Anette Frank

Despite recent successes of large pre-trained language models in solving reasoning tasks, their inference capabilities remain opaque.

Sentence Story Completion

Language Model Decoding as Likelihood-Utility Alignment

1 code implementation13 Oct 2022 Martin Josifoski, Maxime Peyrard, Frano Rajic, Jiheng Wei, Debjit Paul, Valentin Hartmann, Barun Patra, Vishrav Chaudhary, Emre Kiciman, Boi Faltings, Robert West

Specifically, by analyzing the correlation between the likelihood and the utility of predictions across a diverse set of tasks, we provide empirical evidence supporting the proposed taxonomy and a set of principles to structure reasoning when choosing a decoding algorithm.

Language Modelling Text Generation

CRoW: Benchmarking Commonsense Reasoning in Real-World Tasks

1 code implementation23 Oct 2023 Mete Ismayilzada, Debjit Paul, Syrielle Montariol, Mor Geva, Antoine Bosselut

Recent efforts in natural language processing (NLP) commonsense reasoning research have yielded a considerable number of new datasets and benchmarks.

Benchmarking

Social Commonsense Reasoning with Multi-Head Knowledge Attention

1 code implementation Findings of the Association for Computational Linguistics 2020 Debjit Paul, Anette Frank

Notably we are, to the best of our knowledge, the first to demonstrate that a model that learns to perform counterfactual reasoning helps predicting the best explanation in an abductive reasoning task.

counterfactual Counterfactual Reasoning +1

CO-NNECT: A Framework for Revealing Commonsense Knowledge Paths as Explicitations of Implicit Knowledge in Texts

1 code implementation IWCS (ACL) 2021 Maria Becker, Katharina Korfhage, Debjit Paul, Anette Frank

We conduct evaluations on two argumentative datasets and show that a combination of the two model types generates meaningful, high-quality knowledge paths between sentences that reveal implicit knowledge conveyed in text.

Relation

CRAB: Assessing the Strength of Causal Relationships Between Real-world Events

1 code implementation7 Nov 2023 Angelika Romanou, Syrielle Montariol, Debjit Paul, Leo Laugier, Karl Aberer, Antoine Bosselut

In this work, we present CRAB, a new Causal Reasoning Assessment Benchmark designed to evaluate causal understanding of events in real-world narratives.

Generating Hypothetical Events for Abductive Inference

1 code implementation Joint Conference on Lexical and Computational Semantics 2021 Debjit Paul, Anette Frank

This work offers the first study of how such knowledge impacts the Abductive NLI task -- which consists in choosing the more likely explanation for given observations.

Language Modelling

δ-CAUSAL: Exploring Defeasibility in Causal Reasoning

no code implementations6 Jan 2024 Shaobo Cui, Lazar Milikic, Yiyang Feng, Mete Ismayilzada, Debjit Paul, Antoine Bosselut, Boi Faltings

CESAR achieves a significant 69. 7% relative improvement over existing metrics, increasing from 47. 2% to 80. 1% in capturing the causal strength change brought by supporters and defeaters.

Making Reasoning Matter: Measuring and Improving Faithfulness of Chain-of-Thought Reasoning

no code implementations21 Feb 2024 Debjit Paul, Robert West, Antoine Bosselut, Boi Faltings

In this paper, we perform a causal mediation analysis on twelve LLMs to examine how intermediate reasoning steps generated by the LLM influence the final outcome and find that LLMs do not reliably use their intermediate reasoning steps when generating an answer.

counterfactual

Cannot find the paper you are looking for? You can Submit a new open access paper.