On the Noise Robustness of In-Context Learning for Text Generation
- URL: http://arxiv.org/abs/2405.17264v1
- Date: Mon, 27 May 2024 15:22:58 GMT
- Title: On the Noise Robustness of In-Context Learning for Text Generation
- Authors: Hongfu Gao, Feipeng Zhang, Wenyu Jiang, Jun Shu, Feng Zheng, Hongxin Wei,
- Abstract summary: In this work, we show that, on text generation tasks, noisy annotations significantly hurt the performance of in-context learning.
To circumvent the issue, we propose a simple and effective approach called Local Perplexity Ranking (LPR)
LPR replaces the "noisy" candidates with their nearest neighbors that are more likely to be clean.
- Score: 41.59602454113563
- License: http://creativecommons.org/publicdomain/zero/1.0/
- Abstract: Large language models (LLMs) have shown impressive performance on downstream tasks by in-context learning (ICL), which heavily relies on the quality of demonstrations selected from a large set of annotated examples. Recent works claim that in-context learning is robust to noisy demonstrations in text classification. In this work, we show that, on text generation tasks, noisy annotations significantly hurt the performance of in-context learning. To circumvent the issue, we propose a simple and effective approach called Local Perplexity Ranking (LPR), which replaces the "noisy" candidates with their nearest neighbors that are more likely to be clean. Our method is motivated by analyzing the perplexity deviation caused by noisy labels and decomposing perplexity into inherent perplexity and matching perplexity. Our key idea behind LPR is thus to decouple the matching perplexity by performing the ranking among the neighbors in semantic space. Our approach can prevent the selected demonstrations from including mismatched input-label pairs while preserving the effectiveness of the original selection methods. Extensive experiments demonstrate the effectiveness of LPR, improving the EM score by up to 18.75 on common benchmarks with noisy annotations.
Related papers
- Mitigating Boundary Ambiguity and Inherent Bias for Text Classification in the Era of Large Language Models [24.085614720512744]
This study shows that large language models (LLMs) are vulnerable to changes in the number and arrangement of options in text classification.
Key bottleneck arises from ambiguous decision boundaries and inherent biases towards specific tokens and positions.
Our approach is grounded in the empirical observation that pairwise comparisons can effectively alleviate boundary ambiguity and inherent bias.
arXiv Detail & Related papers (2024-06-11T06:53:19Z) - DenoSent: A Denoising Objective for Self-Supervised Sentence
Representation Learning [59.4644086610381]
We propose a novel denoising objective that inherits from another perspective, i.e., the intra-sentence perspective.
By introducing both discrete and continuous noise, we generate noisy sentences and then train our model to restore them to their original form.
Our empirical evaluations demonstrate that this approach delivers competitive results on both semantic textual similarity (STS) and a wide range of transfer tasks.
arXiv Detail & Related papers (2024-01-24T17:48:45Z) - Large Language Models are Efficient Learners of Noise-Robust Speech
Recognition [65.95847272465124]
Recent advances in large language models (LLMs) have promoted generative error correction (GER) for automatic speech recognition (ASR)
In this work, we extend the benchmark to noisy conditions and investigate if we can teach LLMs to perform denoising for GER.
Experiments on various latest LLMs demonstrate our approach achieves a new breakthrough with up to 53.9% correction improvement in terms of word error rate.
arXiv Detail & Related papers (2024-01-19T01:29:27Z) - Noisy Correspondence Learning with Self-Reinforcing Errors Mitigation [63.180725016463974]
Cross-modal retrieval relies on well-matched large-scale datasets that are laborious in practice.
We introduce a novel noisy correspondence learning framework, namely textbfSelf-textbfReinforcing textbfErrors textbfMitigation (SREM)
arXiv Detail & Related papers (2023-12-27T09:03:43Z) - Combating Label Noise With A General Surrogate Model For Sample
Selection [84.61367781175984]
We propose to leverage the vision-language surrogate model CLIP to filter noisy samples automatically.
We validate the effectiveness of our proposed method on both real-world and synthetic noisy datasets.
arXiv Detail & Related papers (2023-10-16T14:43:27Z) - Active Learning Principles for In-Context Learning with Large Language
Models [65.09970281795769]
This paper investigates how Active Learning algorithms can serve as effective demonstration selection methods for in-context learning.
We show that in-context example selection through AL prioritizes high-quality examples that exhibit low uncertainty and bear similarity to the test examples.
arXiv Detail & Related papers (2023-05-23T17:16:04Z) - Compositional Exemplars for In-context Learning [21.961094715261133]
Large pretrained language models (LMs) have shown impressive In-Context Learning (ICL) ability.
We propose CEIL (Compositional Exemplars for In-context Learning) to model the interaction between the given input and in-context examples.
We validate CEIL on 12 classification and generation datasets from 7 distinct NLP tasks, including sentiment analysis, paraphrase detection, natural language inference, commonsense reasoning, open-domain question answering, code generation, and semantic parsing.
arXiv Detail & Related papers (2023-02-11T14:02:08Z) - Improving Few-Shot Performance of Language Models via Nearest Neighbor
Calibration [12.334422701057674]
We propose a novel nearest-neighbor calibration framework for in-context learning.
It is inspired by a phenomenon that the in-context learning paradigm produces incorrect labels when inferring training instances.
Experiments on various few-shot text classification tasks demonstrate that our method significantly improves in-context learning.
arXiv Detail & Related papers (2022-12-05T12:49:41Z) - Context-based Virtual Adversarial Training for Text Classification with
Noisy Labels [1.9508698179748525]
We propose context-based virtual adversarial training (ConVAT) to prevent a text classifier from overfitting to noisy labels.
Unlike the previous works, the proposed method performs the adversarial training at the context level rather than the inputs.
We conduct extensive experiments on four text classification datasets with two types of label noises.
arXiv Detail & Related papers (2022-05-29T14:19:49Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.