Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2402.03686 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929360570679296 |
|---|---|
| author | Sanyal, Soumya Xiao, Tianyi Liu, Jiacheng Wang, Wenya Ren, Xiang |
| author_facet | Sanyal, Soumya Xiao, Tianyi Liu, Jiacheng Wang, Wenya Ren, Xiang |
| contents | Making inferences in text comprehension to understand the meaning is essential in language processing. This work studies the entailment verification (EV) problem of multi-sentence premises that requires a system to make multiple inferences implicitly. Studying EV for such complex premises is important because modern NLP problems, such as detecting inconsistent model-generated rationales, require complex multi-hop reasoning. However, current textual inference datasets mostly contain short premises that only partially focus on these challenges. To address this, we compile an EV benchmark that includes datasets from three NLP domains (NLI, contextual QA, and rationales) containing multi-sentence premises. On benchmarking humans and LLMs, we find that LLMs are better than humans in multi-hop reasoning across extended contexts, while humans perform better in simple deductive reasoning tasks. We also finetune a Flan-T5 model for EV using two training objectives to obtain a strong open-source model that outperforms GPT-3.5 and rivals GPT-4. Finally, we use this model to filter out inconsistent model-generated rationales in self-consistency decoding, resulting in a 6% accuracy improvement on average across three MCQ datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_03686 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Are Machines Better at Complex Reasoning? Unveiling Human-Machine Inference Gaps in Entailment Verification Sanyal, Soumya Xiao, Tianyi Liu, Jiacheng Wang, Wenya Ren, Xiang Computation and Language Artificial Intelligence Making inferences in text comprehension to understand the meaning is essential in language processing. This work studies the entailment verification (EV) problem of multi-sentence premises that requires a system to make multiple inferences implicitly. Studying EV for such complex premises is important because modern NLP problems, such as detecting inconsistent model-generated rationales, require complex multi-hop reasoning. However, current textual inference datasets mostly contain short premises that only partially focus on these challenges. To address this, we compile an EV benchmark that includes datasets from three NLP domains (NLI, contextual QA, and rationales) containing multi-sentence premises. On benchmarking humans and LLMs, we find that LLMs are better than humans in multi-hop reasoning across extended contexts, while humans perform better in simple deductive reasoning tasks. We also finetune a Flan-T5 model for EV using two training objectives to obtain a strong open-source model that outperforms GPT-3.5 and rivals GPT-4. Finally, we use this model to filter out inconsistent model-generated rationales in self-consistency decoding, resulting in a 6% accuracy improvement on average across three MCQ datasets. |
| title | Are Machines Better at Complex Reasoning? Unveiling Human-Machine Inference Gaps in Entailment Verification |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2402.03686 |