"In-Context Learning" or: How I learned to stop worrying and love "Applied Information Retrieval"

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Parry, Andrew, Ganguly, Debasis, Chandra, Manish
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929333567750144
author Parry, Andrew
Ganguly, Debasis
Chandra, Manish
author_facet Parry, Andrew
Ganguly, Debasis
Chandra, Manish
contents With the increasing ability of large language models (LLMs), in-context learning (ICL) has evolved as a new paradigm for natural language processing (NLP), where instead of fine-tuning the parameters of an LLM specific to a downstream task with labeled examples, a small number of such examples is appended to a prompt instruction for controlling the decoder's generation process. ICL, thus, is conceptually similar to a non-parametric approach, such as $k$-NN, where the prediction for each instance essentially depends on the local topology, i.e., on a localised set of similar instances and their labels (called few-shot examples). This suggests that a test instance in ICL is analogous to a query in IR, and similar examples in ICL retrieved from a training set relate to a set of documents retrieved from a collection in IR. While standard unsupervised ranking models can be used to retrieve these few-shot examples from a training set, the effectiveness of the examples can potentially be improved by re-defining the notion of relevance specific to its utility for the downstream task, i.e., considering an example to be relevant if including it in the prompt instruction leads to a correct prediction. With this task-specific notion of relevance, it is possible to train a supervised ranking model (e.g., a bi-encoder or cross-encoder), which potentially learns to optimally select the few-shot examples. We believe that the recent advances in neural rankers can potentially find a use case for this task of optimally choosing examples for more effective downstream ICL predictions.
format Preprint
id arxiv_https___arxiv_org_abs_2405_01116
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle "In-Context Learning" or: How I learned to stop worrying and love "Applied Information Retrieval"
Parry, Andrew
Ganguly, Debasis
Chandra, Manish
Information Retrieval
With the increasing ability of large language models (LLMs), in-context learning (ICL) has evolved as a new paradigm for natural language processing (NLP), where instead of fine-tuning the parameters of an LLM specific to a downstream task with labeled examples, a small number of such examples is appended to a prompt instruction for controlling the decoder's generation process. ICL, thus, is conceptually similar to a non-parametric approach, such as $k$-NN, where the prediction for each instance essentially depends on the local topology, i.e., on a localised set of similar instances and their labels (called few-shot examples). This suggests that a test instance in ICL is analogous to a query in IR, and similar examples in ICL retrieved from a training set relate to a set of documents retrieved from a collection in IR. While standard unsupervised ranking models can be used to retrieve these few-shot examples from a training set, the effectiveness of the examples can potentially be improved by re-defining the notion of relevance specific to its utility for the downstream task, i.e., considering an example to be relevant if including it in the prompt instruction leads to a correct prediction. With this task-specific notion of relevance, it is possible to train a supervised ranking model (e.g., a bi-encoder or cross-encoder), which potentially learns to optimally select the few-shot examples. We believe that the recent advances in neural rankers can potentially find a use case for this task of optimally choosing examples for more effective downstream ICL predictions.
title "In-Context Learning" or: How I learned to stop worrying and love "Applied Information Retrieval"
topic Information Retrieval
url https://arxiv.org/abs/2405.01116