paper-with-me

홈 › Papers

Selective Self-to-Supervised Fine-Tuning for Generalization in Large Language Models

2025-02-12 · Sonam Gupta, Yatin Nandwani, Asaf Yehudai, Dinesh Khandelwal, Dinesh Raghu, Sachindra Joshi

Fine-tuning Large Language Models (LLMs) on specific datasets is a common practice to improve performance on target tasks. However, this performance gain often leads to overfitting, where the model becomes too specialized in either the task or the characteristics of the training data, resulting in a loss of generalization. This paper introduces Selective Self-to-Supervised Fine-Tuning (S3FT), a fine-tuning approach that achieves better performance than the standard supervised fine-tuning (SFT) while improving generalization. S3FT leverages the existence of multiple valid responses to a query. By utilizing the model's correct responses, S3FT reduces model specialization during the fine-tuning stage. S3FT first identifies the correct model responses from the training set by deploying an appropriate judge. Then, it fine-tunes the model using the correct model responses and the gold response (or its paraphrase) for the remaining samples. The effectiveness of S3FT is demonstrated through experiments on mathematical reasoning, Python programming and reading comprehension tasks. The results show that standard SFT can lead to an average performance drop of up to $4.4$ on multiple benchmarks, such as MMLU and TruthfulQA. In contrast, S3FT reduces this drop by half, i.e. $2.5$, indicating better generalization capabilities than SFT while performing significantly better on the fine-tuning tasks.

📄 PDF Abstract BibTeX arXiv:2502.08130

Code (0)

등록된 구현이 없습니다.

Tasks

Mathematical ReasoningMMLUReading ComprehensionTruthfulQA

Methods 이 논문이 사용한 방법론

SFT Shrink and Fine-Tune, or SFT, is a type of distillation that avoids explicit distillation by copying parameters to a student student model and then fine-tuning.…
SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

Selective Self-Rehearsal: A Fine-Tuning Approach to Improve Generalization in Large Language Models

2024-09-07 · Sonam Gupta, Yatin Nandwani, Asaf Yehudai, Mayank Mishra 외

Fine-tuning Large Language Models (LLMs) on specific datasets is a common practice to improve performance on target tasks. However, this performance gain often leads to overfitting, where the model becomes too specialize…

MMLUTruthfulQAvalid

Windowed SummaryMixing: An Efficient Fine-Tuning of Self-Supervised Learning Models for Low-resource Speech Recognition

2026-02-04 · Aditya Srinivas Menon, Kumud Tripathi, Raj Gohil, Pankaj Wasnik arxiv

Self-supervised learning (SSL) has advanced speech processing but suffers from quadratic complexity due to self-attention. To address this, SummaryMixing (SM) has been proposed as a linear-time alternative that summarize…

Self-Supervised LearningSpeech Recognition

Improving Adversarial Defense with Self-supervised Test-time Fine-tuning

2021-09-29 · Zhichao Huang, Chen Liu, Mathieu Salzmann, Sabine Süsstrunk 외

Although adversarial training and its variants currently constitute the most effective way to achieve robustness against adversarial attacks, their poor generalization limits their performance on the test samples. In thi…

Adversarial Defense

Less Forgetting for Better Generalization: Exploring Continual-learning Fine-tuning Methods for Speech Self-supervised Representations

2024-06-30 · Salah Zaiem, Titouan Parcollet, Slim Essid

Despite being trained on massive and diverse datasets, speech self-supervised encoders are generally used for downstream purposes as mere frozen feature extractors or model initializers before fine-tuning. The former sev…

Continual LearningDomain Generalizationspeech-recognitionSpeech Recognition

Learn from Downstream and Be Yourself in Multimodal Large Language Model Fine-Tuning

2024-11-17 · Wenke Huang, Jian Liang, Zekun Shi, Didi Zhu 외

Multimodal Large Language Model (MLLM) have demonstrated strong generalization capabilities across diverse distributions and tasks, largely due to extensive pre-training datasets. Fine-tuning MLLM has become a common pra…

Image CaptioningLanguage ModelingLanguage ModellingLarge Language Model+3