RobustFT: Robust Supervised Fine-tuning for Large Language Models under Noisy Response

• URL: http://arxiv.org/abs/2412.14922v1
• Date: Thu, 19 Dec 2024 15:00:18 GMT
• Title: RobustFT: Robust Supervised Fine-tuning for Large Language Models under Noisy Response
• Authors: Junyu Luo, Xiao Luo, Kaize Ding, Jingyang Yuan, Zhiping Xiao, Ming Zhang,
• Abstract summary: Supervised fine-tuning (SFT) plays a crucial role in adapting large language models (LLMs) to specific domains or tasks. We introduce a robust SFT framework (RobustFT) that performs noise detection and relabeling on downstream task data. Extensive experiments conducted on multiple LLMs across five datasets demonstrate RobustFT's exceptional performance in noisy scenarios.
• Score: 23.45168175163634
• License:
• Abstract: Supervised fine-tuning (SFT) plays a crucial role in adapting large language models (LLMs) to specific domains or tasks. However, as demonstrated by empirical experiments, the collected data inevitably contains noise in practical applications, which poses significant challenges to model performance on downstream tasks. Therefore, there is an urgent need for a noise-robust SFT framework to enhance model capabilities in downstream tasks. To address this challenge, we introduce a robust SFT framework (RobustFT) that performs noise detection and relabeling on downstream task data. For noise identification, our approach employs a multi-expert collaborative system with inference-enhanced models to achieve superior noise detection. In the denoising phase, we utilize a context-enhanced strategy, which incorporates the most relevant and confident knowledge followed by careful assessment to generate reliable annotations. Additionally, we introduce an effective data selection mechanism based on response entropy, ensuring only high-quality samples are retained for fine-tuning. Extensive experiments conducted on multiple LLMs across five datasets demonstrate RobustFT's exceptional performance in noisy scenarios.

Related papers

• Do we really have to filter out random noise in pre-training data for language models? [42.966566701950164]
  Pre-training text data curated from the internet inevitably contains random noise caused by decoding errors or unregulated web content. We provide a theoretical justification for this phenomenon, which also elucidates the success of multilingual models. Experiments show that the model's performance in downstream tasks is not based solely on the NTP loss, which means that random noise may result in degraded downstream performance. We introduce a novel plug-and-play Local Gradient Matching loss, which explicitly enhances the denoising capability of the downstream task head.
  arXiv  Detail & Related papers  (2025-02-10T16:01:55Z)
• One-step Noisy Label Mitigation [86.57572253460125]
  Mitigating the detrimental effects of noisy labels on the training process has become increasingly critical. We propose One-step Anti-Noise (OSA), a model-agnostic noisy label mitigation paradigm. We empirically demonstrate the superiority of OSA, highlighting its enhanced training robustness, improved task transferability, ease of deployment, and reduced computational costs.
  arXiv  Detail & Related papers  (2024-10-02T18:42:56Z)
• Towards Building a Robust Knowledge Intensive Question Answering Model with Large Language Models [4.4849006637642805]
  Presence of noise and errors in retrieved information poses challenges to the robustness of LLMs. To address the issue of model accuracy decline caused by noisy external information, we propose a data augmentation-based fine-tuning method. We have conducted experiments on both existing LLMs and our approach, the results are evaluated by GPT-4.
  arXiv  Detail & Related papers  (2024-09-09T07:32:30Z)
• Effective Noise-aware Data Simulation for Domain-adaptive Speech Enhancement Leveraging Dynamic Stochastic Perturbation [25.410770364140856]
  Cross-domain speech enhancement (SE) is often faced with severe challenges due to the scarcity of noise and background information in an unseen target domain. This study puts forward a novel data simulation method to address this issue, leveraging noise-extractive techniques and generative adversarial networks (GANs) We introduce the notion of dynamic perturbation, which can inject controlled perturbations into the noise embeddings during inference.
  arXiv  Detail & Related papers  (2024-09-03T02:29:01Z)
• ROPO: Robust Preference Optimization for Large Language Models [59.10763211091664]
  We propose an iterative alignment approach that integrates noise-tolerance and filtering of noisy samples without the aid of external models. Experiments on three widely-used datasets with Mistral-7B and Llama-2-7B demonstrate that ROPO significantly outperforms existing preference alignment methods.
  arXiv  Detail & Related papers  (2024-04-05T13:58:51Z)
• Learning with Noisy Foundation Models [95.50968225050012]
  This paper is the first work to comprehensively understand and analyze the nature of noise in pre-training datasets. We propose a tuning method (NMTune) to affine the feature space to mitigate the malignant effect of noise and improve generalization.
  arXiv  Detail & Related papers  (2024-03-11T16:22:41Z)
• Revisit Input Perturbation Problems for LLMs: A Unified Robustness Evaluation Framework for Noisy Slot Filling Task [18.623619585980688]
  We propose a unified robustness evaluation framework based on the slot-filling task to evaluate the dialogue understanding capability of large language models. Specifically, we construct a input perturbation evaluation dataset, Noise-LLM, which contains five types of single perturbation and four types of mixed perturbation data. Our aim is to assess how well various robustness methods of LLMs perform in real-world noisy scenarios.
  arXiv  Detail & Related papers  (2023-10-10T10:22:05Z)
• Towards Robust and Generalizable Training: An Empirical Study of Noisy Slot Filling for Input Perturbations [38.766702041991046]
  We introduce a noise robustness evaluation dataset named Noise-SF for slot filling task. The proposed dataset contains five types of human-annotated noise. We find that baseline models have poor performance in robustness evaluation.
  arXiv  Detail & Related papers  (2023-10-05T12:59:57Z)
• DiffSED: Sound Event Detection with Denoising Diffusion [70.18051526555512]
  We reformulate the SED problem by taking a generative learning perspective. Specifically, we aim to generate sound temporal boundaries from noisy proposals in a denoising diffusion process. During training, our model learns to reverse the noising process by converting noisy latent queries to the groundtruth versions.
  arXiv  Detail & Related papers  (2023-08-14T17:29:41Z)
• Improve Noise Tolerance of Robust Loss via Noise-Awareness [60.34670515595074]
  We propose a meta-learning method which is capable of adaptively learning a hyper parameter prediction function, called Noise-Aware-Robust-Loss-Adjuster (NARL-Adjuster for brevity) Four SOTA robust loss functions are attempted to be integrated with our algorithm, and comprehensive experiments substantiate the general availability and effectiveness of the proposed method in both its noise tolerance and performance.
  arXiv  Detail & Related papers  (2023-01-18T04:54:58Z)
• Improving the Robustness of Summarization Models by Detecting and Removing Input Noise [50.27105057899601]
  We present a large empirical study quantifying the sometimes severe loss in performance from different types of input noise for a range of datasets and model sizes. We propose a light-weight method for detecting and removing such noise in the input during model inference without requiring any training, auxiliary models, or even prior knowledge of the type of noise.
  arXiv  Detail & Related papers  (2022-12-20T00:33:11Z)

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.