Lost in Inference: Rediscovering the Role of Natural Language Inference for Large Language Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910707275005952 |
|---|---|
| author | Madaan, Lovish Esiobu, David Stenetorp, Pontus Plank, Barbara Hupkes, Dieuwke |
| author_facet | Madaan, Lovish Esiobu, David Stenetorp, Pontus Plank, Barbara Hupkes, Dieuwke |
| contents | In the recent past, a popular way of evaluating natural language understanding (NLU), was to consider a model's ability to perform natural language inference (NLI) tasks. In this paper, we investigate if NLI tasks, that are rarely used for LLM evaluation, can still be informative for evaluating LLMs. Focusing on five different NLI benchmarks across six models of different scales, we investigate if they are able to discriminate models of different size and quality and how their accuracies develop during training. Furthermore, we investigate the extent to which the softmax distributions of models align with human distributions in cases where statements are ambiguous or vague. Overall, our results paint a positive picture for the NLI tasks: we find that they are able to discriminate well between models at various stages of training, yet are not (all) saturated. Furthermore, we find that while the similarity of model distributions with human label distributions increases with scale, it is still much higher than the similarity between two populations of humans, making it a potentially interesting statistic to consider. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_14103 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Lost in Inference: Rediscovering the Role of Natural Language Inference for Large Language Models Madaan, Lovish Esiobu, David Stenetorp, Pontus Plank, Barbara Hupkes, Dieuwke Computation and Language In the recent past, a popular way of evaluating natural language understanding (NLU), was to consider a model's ability to perform natural language inference (NLI) tasks. In this paper, we investigate if NLI tasks, that are rarely used for LLM evaluation, can still be informative for evaluating LLMs. Focusing on five different NLI benchmarks across six models of different scales, we investigate if they are able to discriminate models of different size and quality and how their accuracies develop during training. Furthermore, we investigate the extent to which the softmax distributions of models align with human distributions in cases where statements are ambiguous or vague. Overall, our results paint a positive picture for the NLI tasks: we find that they are able to discriminate well between models at various stages of training, yet are not (all) saturated. Furthermore, we find that while the similarity of model distributions with human label distributions increases with scale, it is still much higher than the similarity between two populations of humans, making it a potentially interesting statistic to consider. |
| title | Lost in Inference: Rediscovering the Role of Natural Language Inference for Large Language Models |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2411.14103 |