Disperse-Then-Merge: Pushing the Limits of Instruction Tuning via Alignment Tax Reduction

• URL: http://arxiv.org/abs/2405.13432v1
• Date: Wed, 22 May 2024 08:18:19 GMT
• Title: Disperse-Then-Merge: Pushing the Limits of Instruction Tuning via Alignment Tax Reduction
• Authors: Tingchen Fu, Deng Cai, Lemao Liu, Shuming Shi, Rui Yan,
• Abstract summary: Large language models (LLMs) tend to suffer from deterioration at the latter stage ofSupervised fine-tuning process. We introduce a simple disperse-then-merge framework to address the issue. Our framework outperforms various sophisticated methods such as data curation and training regularization on a series of standard knowledge and reasoning benchmarks.
• Score: 75.25114727856861
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Supervised fine-tuning (SFT) on instruction-following corpus is a crucial approach toward the alignment of large language models (LLMs). However, the performance of LLMs on standard knowledge and reasoning benchmarks tends to suffer from deterioration at the latter stage of the SFT process, echoing the phenomenon of alignment tax. Through our pilot study, we put a hypothesis that the data biases are probably one cause behind the phenomenon. To address the issue, we introduce a simple disperse-then-merge framework. To be concrete, we disperse the instruction-following data into portions and train multiple sub-models using different data portions. Then we merge multiple models into a single one via model merging techniques. Despite its simplicity, our framework outperforms various sophisticated methods such as data curation and training regularization on a series of standard knowledge and reasoning benchmarks.

Related papers

• Context-Parametric Inversion: Why Instruction Finetuning May Not Actually Improve Context Reliance [68.56701216210617]
  In-principle, one would expect models to adapt to the user context better after instruction finetuning. We observe a surprising failure mode: during instruction tuning, the context reliance under knowledge conflicts initially increases as expected, but then gradually decreases.
  arXiv  Detail & Related papers  (2024-10-14T17:57:09Z)
• Reference Trustable Decoding: A Training-Free Augmentation Paradigm for Large Language Models [79.41139393080736]
  Large language models (LLMs) have rapidly advanced and demonstrated impressive capabilities. We propose Reference Trustable Decoding (RTD), a paradigm that allows models to quickly adapt to new tasks without fine-tuning.
  arXiv  Detail & Related papers  (2024-09-30T10:48:20Z)
• Fine-Tuning with Divergent Chains of Thought Boosts Reasoning Through Self-Correction in Language Models [63.36637269634553]
  We present a novel method of further improving performance by requiring models to compare multiple reasoning chains. We find that instruction tuning on DCoT datasets boosts the performance of even smaller, and therefore more accessible, language models.
  arXiv  Detail & Related papers  (2024-07-03T15:01:18Z)
• Can LLMs Separate Instructions From Data? And What Do We Even Mean By That? [60.50127555651554]
  Large Language Models (LLMs) show impressive results in numerous practical applications, but they lack essential safety features. This makes them vulnerable to manipulations such as indirect prompt injections and generally unsuitable for safety-critical tasks. We introduce a formal measure for instruction-data separation and an empirical variant that is calculable from a model's outputs.
  arXiv  Detail & Related papers  (2024-03-11T15:48:56Z)
• Self-Distillation Bridges Distribution Gap in Language Model Fine-Tuning [29.72531930169291]
  Self-Distillation Fine-Tuning (SDFT) is a novel approach that bridges the distribution gap by guiding fine-tuning with a distilled dataset generated by the model itself to match its original distribution. Experimental results on the Llama-2-chat model across various benchmarks demonstrate that SDFT effectively mitigates catastrophic forgetting.
  arXiv  Detail & Related papers  (2024-02-21T10:06:08Z)
• Becoming self-instruct: introducing early stopping criteria for minimal instruct tuning [0.0]
  We introduce the Instruction Following Score (IFS), a metric that detects language models' ability to follow instructions. We benchmark publicly available base and instruct models, and show that the ratio of well formatted responses to partial and full sentences can be an effective measure. We compute IFS for Supervised Fine-Tuning (SFT) of 7B and 13B LLaMA models, showing that models learn to follow instructions relatively early in the training process.
  arXiv  Detail & Related papers  (2023-07-05T09:42:25Z)
• MaxMatch: Semi-Supervised Learning with Worst-Case Consistency [149.03760479533855]
  We propose a worst-case consistency regularization technique for semi-supervised learning (SSL) We present a generalization bound for SSL consisting of the empirical loss terms observed on labeled and unlabeled training data separately. Motivated by this bound, we derive an SSL objective that minimizes the largest inconsistency between an original unlabeled sample and its multiple augmented variants.
  arXiv  Detail & Related papers  (2022-09-26T12:04:49Z)
• Few-shot Learning via Dependency Maximization and Instance Discriminant Analysis [21.8311401851523]
  We study the few-shot learning problem, where a model learns to recognize new objects with extremely few labeled data per category. We propose a simple approach to exploit unlabeled data accompanying the few-shot task for improving few-shot performance.
  arXiv  Detail & Related papers  (2021-09-07T02:19:01Z)

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.