SFTMix: Elevating Language Model Instruction Tuning with Mixup Recipe (2026.acl-long)
Copied to clipboard
| Challenge: | Efforts to improve instruction tuning often focus on higher-quality supervised fine-tuning datasets, typically requiring data filtering with proprietary LLMs or human annotation. |
| Approach: | They propose a Mixup-based recipe that elevates LLM instruction tuning without relying on well-curated datasets. |
| Outcome: | The proposed model improves instruction-following and healthcare-specific tasks with consistent improvements across LLM families and SFT datasets. |
Similar Papers
Demystifying Instruction Mixing for Fine-tuning Large Language Models (2024.acl-srw)
Copied to clipboard
| Challenge: | Instruction tuning is effective for aligning large language models with human instructions, but the procedure to optimizing the mixing of instruction datasets is still unclear. |
| Approach: | They categorize instructions into three primary types: NLP downstream tasks, coding, and general chat. |
| Outcome: | The proposed method improves performance of large language models (LLMs) but it is difficult to combine different instruction datasets to optimize overall performance. |
Massive Supervised Fine-tuning Experiments Reveal How Data, Layer, and Training Factors Shape LLM Alignment Quality (2025.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in large language models (LLMs) have greatly improved natural language understanding and generation. |
| Approach: | They train a wide range of base models on a variety of datasets including code generation, mathematical reasoning, and general-domain tasks. |
| Outcome: | The results show that training–task synergies persist across all models while others vary substantially, emphasizing the importance of model-specific strategies. |
From English to Second Language Mastery: Enhancing LLMs with Cross-Lingual Continued Instruction Tuning (2025.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) acquire strong language skills through extensive pre-training and supervised fine-tuning (SFT) on instructionresponse pairs. |
| Approach: | They propose a method which leverages translation-based parallel instruction data to enhance cross-lingual adaptability. |
| Outcome: | The proposed model improves on Llama-2-7B across five languages against three objective benchmarks and an LLM-as-a-judge benchmark. |
RECOST: External Knowledge Guided Data-efficient Instruction Tuning (2024.findings-acl)
Copied to clipboard
| Challenge: | Considering the high computing power overhead, data-efficient instruction tuning is proposed to reduce the training data size. |
| Approach: | They propose a framework to improve instruction tuning by integrating external knowledge into a single pipeline. |
| Outcome: | The proposed method achieves better results with only 1% of the full dataset. |
Tuna: Instruction Tuning using Feedback from Large Language Models (2023.findings-emnlp)
Copied to clipboard
| Challenge: | LLms like LLaMA have shown to be cost-effective for generating better responses . however, the instruction-tuned model has only seen one response per instruction . |
| Approach: | They propose to fine tune an instruction-tuned LLM using probabilistic ranking and contextual ranking approaches to increase the likelihood of generating better responses. |
| Outcome: | The proposed model improves on Super Natural Instructions, LMentry and Vicuna QA. |
CodecLM: Aligning Language Models with Tailored Synthetic Data (2024.findings-naacl)
Copied to clipboard
Zifeng Wang, Chun-Liang Li, Vincent Perot, Long Le, Jin Miao, Zizhao Zhang, Chen-Yu Lee, Tomas Pfister
| Challenge: | Recent work on generating diverse instructions and applying LLM to increase instruction complexity neglects downstream use cases. |
| Approach: | They propose a framework for generating high-quality synthetic data for LLM alignment with different downstream instruction distributions and LLMs. |
| Outcome: | Experiments on four open-domain instruction using the proposed framework validate the effectiveness of CodecLM over the current state-of-the-art. |
Semi-supervised Fine-tuning for Large Language Models (2025.findings-naacl)
Copied to clipboard
| Challenge: | Existing LLMs require labeled data, which can be costly in real-world applications. |
| Approach: | They propose a framework that can fully exploit labeled and unlabeled data for LLM fine-tuning . they conducted experiments using GPT-4o-mini and Llama-3.1 on seven general or domain-specific datasets . |
| Outcome: | The proposed framework can fully exploit labeled and unlabeled data for LLM alignment from a propagate-and-select manner. |
Low-Confidence Gold: Refining Low-Confidence Samples for Efficient Instruction Tuning (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Low-Confidence Gold (LCG) is a new filtering framework for Large Language Models that curates high-quality subsets while preserving data diversity. |
| Approach: | They propose a new filtering framework that employs centroid-based clustering and confidence-guided selection for identifying valuable instruction pairs. |
| Outcome: | The proposed framework improves performance on a subset of 6K samples while maintaining data diversity. |
Disperse-Then-Merge: Pushing the Limits of Instruction Tuning via Alignment Tax Reduction (2024.findings-acl)
Copied to clipboard
| Challenge: | Pre-trained language models may not follow human instructions and produce toxic, hallucinated, or biased content. |
| Approach: | They propose a disperse-then-merge framework that dispersers instruction-following data into portions and trains multiple sub-models using different data portions. |
| Outcome: | The proposed framework outperforms data curation and training regularization on standard knowledge and reasoning benchmarks. |
CommonIT: Commonality-Aware Instruction Tuning for Large Language Models via Data Partitions (2024.emnlp-main)
Copied to clipboard
| Challenge: | Current studies have focused on fine-tuning, but the use of instruction tuning is not as effective as fine-cuning. |
| Approach: | They propose a commonality-aware instruction tuning strategy to cluster instruction datasets into distinct groups with three proposed metrics Task, Embedding and Length. |
| Outcome: | The proposed strategy boosts an average improvement of 2.1% on the general domain and 5.2% on the special domain. |