From 6431feefe6debb530a98d9aaeef2084c67f4497b Mon Sep 17 00:00:00 2001 From: Shreya Havaldar Date: Wed, 15 Jan 2025 11:54:21 -0500 Subject: [PATCH 1/3] added paper link and citation --- README.md | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/README.md b/README.md index 4594577..9d82ba4 100644 --- a/README.md +++ b/README.md @@ -1,5 +1,7 @@ # Implied NLI +Code & resources for the paper [Entailed Between the Lines: Incorporating Implication into NLI](https://arxiv.org/abs/2501.07719) + ### Dataset Overview **Implied NLI**, or **INLI** is a corpus of 10k premises that mirror real-world communication and 40k hypotheses that are implied, explicit, neutral, and contradictory. @@ -8,7 +10,7 @@ On top of the raw data, we also include all prompts used to generate INLI, all human annotations described in the paper, and the outputs of the LLM experiments described in the paper. -For more details on the design and content of the dataset, please see [our paper]() +For more details on the design and content of the dataset, please see [our paper](https://arxiv.org/abs/2501.07719) ### Data Format Our raw dataset is split into three csv files. @@ -27,6 +29,17 @@ In each of these files, one row represents a premise and four corresponding hypo * `contradiction`: A hypothesis contradicted by the premise ## Citation +``` +@misc{havaldar2025entailedlinesincorporatingimplication, + title={Entailed Between the Lines: Incorporating Implication into NLI}, + author={Shreya Havaldar and Hamidreza Alvari and Alex Fabrikant and John Palowitch and Mohammad Javad Hosseini and Senaka Buthpitiya}, + year={2025}, + eprint={2501.07719}, + archivePrefix={arXiv}, + primaryClass={cs.CL}, + url={https://arxiv.org/abs/2501.07719}, +} +``` ## Contact From 98c740893445aee07ace38877237595ea7a9d4fb Mon Sep 17 00:00:00 2001 From: Shreya Havaldar Date: Wed, 15 Jan 2025 11:58:53 -0500 Subject: [PATCH 2/3] added paper link and citation --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 9d82ba4..1ff5ac6 100644 --- a/README.md +++ b/README.md @@ -6,7 +6,7 @@ Code & resources for the paper [Entailed Between the Lines: Incorporating Implic **Implied NLI**, or **INLI** is a corpus of 10k premises that mirror real-world communication and 40k hypotheses that are implied, explicit, neutral, and contradictory. * Number of examples: 40,000 -* Number of labels: 4 (implied entailment entailment, explicit, neutral, contradictory) +* Number of labels: 4 (implied entailment, explicit entailment, neutral, contradictory) On top of the raw data, we also include all prompts used to generate INLI, all human annotations described in the paper, and the outputs of the LLM experiments described in the paper. From d07a9a12ae6487fff455d98030fc2239345c66b7 Mon Sep 17 00:00:00 2001 From: Shreya Havaldar Date: Thu, 16 Jan 2025 15:08:05 -0500 Subject: [PATCH 3/3] testing email chnage --- README.md | 1 - 1 file changed, 1 deletion(-) diff --git a/README.md b/README.md index 1ff5ac6..eb45f1c 100644 --- a/README.md +++ b/README.md @@ -42,7 +42,6 @@ In each of these files, one row represents a premise and four corresponding hypo ``` ## Contact - [Shreya Havaldar](https://shreyahavaldar.com/) ## Disclaimer