Correcting Large Language Model Behavior via Influence Function

• URL: http://arxiv.org/abs/2412.16451v1
• Date: Sat, 21 Dec 2024 02:50:08 GMT
• Title: Correcting Large Language Model Behavior via Influence Function
• Authors: Han Zhang, Zhuo Zhang, Yi Zhang, Yuanzhao Zhai, Hanyang Peng, Yu Lei, Yue Yu, Hui Wang, Bin Liang, Lin Gui, Ruifeng Xu,
• Abstract summary: The dynamic nature of human preferences can render some prior training data outdated or even erroneous. We propose a novel approach, Large Language Model Behavior Correction with Influence Function Recall and Post-Training (LANCET) LANCET consists of two phases: (1) using influence functions to identify the training data that significantly impact undesirable model outputs, and (2) applying an Influence function-driven Bregman Optimization (IBO) technique to adjust the model's behavior.
• Score: 44.090990384733324
• License:
• Abstract: Recent advancements in AI alignment techniques have significantly improved the alignment of large language models (LLMs) with static human preferences. However, the dynamic nature of human preferences can render some prior training data outdated or even erroneous, ultimately causing LLMs to deviate from contemporary human preferences and societal norms. Existing methodologies, whether they involve the curation of new data for continual alignment or the manual correction of outdated data for re-alignment, demand costly human resources. To address this challenge, we propose a novel approach, Large Language Model Behavior Correction with Influence Function Recall and Post-Training (LANCET), which requires no human involvement. LANCET consists of two phases: (1) using influence functions to identify the training data that significantly impact undesirable model outputs, and (2) applying an Influence function-driven Bregman Optimization (IBO) technique to adjust the model's behavior based on these influence distributions. Our experiments demonstrate that LANCET effectively and efficiently correct inappropriate behaviors of LLMs. Furthermore, LANCET can outperform methods that rely on collecting human preferences, and it enhances the interpretability of learning human preferences within LLMs.

Related papers

• Dynamic Loss-Based Sample Reweighting for Improved Large Language Model Pretraining [55.262510814326035]
  Existing reweighting strategies primarily focus on group-level data importance. We introduce novel algorithms for dynamic, instance-level data reweighting. Our framework allows us to devise reweighting strategies deprioritizing redundant or uninformative data.
  arXiv  Detail & Related papers  (2025-02-10T17:57:15Z)
• Optimizing LLMs with Direct Preferences: A Data Efficiency Perspective [4.548047308860141]
  This study investigates the impact of different type of preference data on model performance. It aims to reduce their dependency on extensive amounts of preference data, which is expensive to collect.
  arXiv  Detail & Related papers  (2024-10-22T00:11:41Z)
• Enhancing Training Data Attribution for Large Language Models with Fitting Error Consideration [74.09687562334682]
  We introduce a novel training data attribution method called Debias and Denoise Attribution (DDA) Our method significantly outperforms existing approaches, achieving an averaged AUC of 91.64%. DDA exhibits strong generality and scalability across various sources and different-scale models like LLaMA2, QWEN2, and Mistral.
  arXiv  Detail & Related papers  (2024-10-02T07:14:26Z)
• HARE: HumAn pRiors, a key to small language model Efficiency [6.253561984966316]
  Human priors play a crucial role in efficiently utilizing data in deep learning. Existing Small Language Models mainly rely on web-scraped large-scale training data. We propose a principle to leverage human priors for data construction.
  arXiv  Detail & Related papers  (2024-06-17T10:56:03Z)
• Getting More Juice Out of the SFT Data: Reward Learning from Human Demonstration Improves SFT for LLM Alignment [65.15914284008973]
  We propose to leverage an Inverse Reinforcement Learning (IRL) technique to simultaneously build an reward model and a policy model. We show that the proposed algorithms converge to the stationary solutions of the IRL problem. Our results indicate that it is beneficial to leverage reward learning throughout the entire alignment process.
  arXiv  Detail & Related papers  (2024-05-28T07:11:05Z)
• CLHA: A Simple yet Effective Contrastive Learning Framework for Human Alignment [42.71324708567498]
  Reinforcement learning from human feedback (RLHF) is a crucial technique in aligning large language models (LLMs) with human preferences. We present a simple yet effective Contrastive Learning Framework for Human Alignment (CLHA) to align LLMs with human preferences directly.
  arXiv  Detail & Related papers  (2024-03-25T11:37:15Z)
• Secrets of RLHF in Large Language Models Part II: Reward Modeling [134.97964938009588]
  We introduce a series of novel methods to mitigate the influence of incorrect and ambiguous preferences in the dataset. We also introduce contrastive learning to enhance the ability of reward models to distinguish between chosen and rejected responses.
  arXiv  Detail & Related papers  (2024-01-11T17:56:59Z)
• SALMON: Self-Alignment with Instructable Reward Models [80.83323636730341]
  This paper presents a novel approach, namely SALMON, to align base language models with minimal human supervision. We develop an AI assistant named Dromedary-2 with only 6 exemplars for in-context learning and 31 human-defined principles.
  arXiv  Detail & Related papers  (2023-10-09T17:56:53Z)
• Federated Learning for Early Dropout Prediction on Healthy Ageing Applications [0.0]
  We present a federated machine learning (FML) approach that minimizes privacy concerns and enables distributed training, without transferring individual data. Our results show that data selection and class imbalance handling techniques significantly improve the predictive accuracy of models trained under FML.
  arXiv  Detail & Related papers  (2023-09-08T13:17:06Z)

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.