# DynamixSFT: Dynamic Mixture Optimization of Instruction Tuning Collections

arXiv: https://arxiv.org/abs/2508.12116 · HTML: https://mlsanity.com/paper/2508.12116/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 42 references that are in the map.
- First posted on arXiv 16 Aug 2025.

## Builds on (most cited first, 30 of 34)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them](https://mlsanity.com/paper/2210.09261/) (2022)
- [Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation](https://mlsanity.com/paper/2305.01210/) (2023)
- [When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memories](https://mlsanity.com/paper/2212.10511/) (2022)
- [WizardLM: Empowering large pre-trained language models to follow complex instructions](https://mlsanity.com/paper/2304.12244/) (2023)
- [LIMA: Less Is More for Alignment](https://mlsanity.com/paper/2305.11206/) (2023)
- [Tulu 3: Pushing Frontiers in Open Language Model Post-Training](https://mlsanity.com/paper/2411.15124/) (2024)
- [DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs](https://mlsanity.com/paper/1903.00161/) (2019)
- [The Flan Collection: Designing Data and Methods for Effective Instruction Tuning](https://mlsanity.com/paper/2301.13688/) (2023)
- [WizardCoder: Empowering Code Large Language Models with Evol-Instruct](https://mlsanity.com/paper/2306.08568/) (2023)
- [OpenAssistant Conversations -- Democratizing Large Language Model Alignment](https://mlsanity.com/paper/2304.07327/) (2023)
- [WildChat: 1M ChatGPT Interaction Logs in the Wild](https://mlsanity.com/paper/2405.01470/) (2024)
- [Fact or Fiction: Verifying Scientific Claims](https://mlsanity.com/paper/2004.14974/) (2020)
- [A Dataset of Information-Seeking Questions and Answers Anchored in Research Papers](https://mlsanity.com/paper/2105.03011/) (2021)
- [How Far Can Camels Go? Exploring the State of Instruction Tuning on Open Resources](https://mlsanity.com/paper/2306.04751/) (2023)
- [DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretraining](https://mlsanity.com/paper/2305.10429/) (2023)
- [Multi-Task Identification of Entities, Relations, and Coreference for Scientific Knowledge Graph Construction](https://mlsanity.com/paper/1808.09602/) (2018)
- [OpenMathInstruct-2: Accelerating AI for Math with Massive Open-Source Instruction Data](https://mlsanity.com/paper/2410.01560/) (2024)
- [RegMix: Data Mixture as Regression for Language Model Pre-training](https://mlsanity.com/paper/2407.01492/) (2024)
- [TLDR: Extreme Summarization of Scientific Documents](https://mlsanity.com/paper/2004.15011/) (2020)
- [Self-Evolving Curriculum for LLM Reasoning](https://mlsanity.com/paper/2505.14970/) (2025)
- [The Art of Saying No: Contextual Noncompliance in Language Models](https://mlsanity.com/paper/2407.12043/) (2024)
- [DoGE: Domain Reweighting with Generalization Estimation](https://mlsanity.com/paper/2310.15393/) (2023)
- [Efficient Online Data Mixing For Language Model Pre-Training](https://mlsanity.com/paper/2312.02406/) (2023)
- [Balancing Training for Multilingual Neural Machine Translation](https://mlsanity.com/paper/2004.06748/) (2020)
- [Boltzmann Exploration Done Right](https://mlsanity.com/paper/1705.10257/) (2017)

## Built on by (most cited first, 9 of 9)

- [DynaMiCS: Fine-tuning LLMs with Performance Constraints using Dynamic Mixtures](https://mlsanity.com/paper/2605.10770/) (2026)
- [mSFT: Addressing Dataset Mixtures Overfitting Heterogeneously in Multi-task SFT](https://mlsanity.com/paper/2603.21606/) (2026)
- [A Component-Based Survey of Interactions between Large Language Models and Multi-Armed Bandits](https://mlsanity.com/paper/2601.12945/) (2026)
- [Beyond Full-Model Rollback: AuroSFT for Adapter-State Multi-Task Fine-Tuning](https://mlsanity.com/paper/2608.05250/) (2026)
- [How code helps different tasks? A decompositional lens on LLM post-training](https://mlsanity.com/paper/2609.33845/) (2026)
- [CALIBUDGET: Calibration-Guided Source Allocation for Fixed-Budget Mixed-Reasoning Adaptation](https://mlsanity.com/paper/2609.36721/) (2026)
- [Adaptive Mutual Distillation for Balanced Multi-Task Post-Training of Large Language Models](https://mlsanity.com/paper/2610.02856/) (2026)
- [Dynamic Minimax Regret Optimization for Robust LLM Post-Training](https://mlsanity.com/paper/2610.06329/) (2026)
- [LLM Agents for Automated Post-Training via Structured Recipe Optimization](https://mlsanity.com/paper/2610.18572/) (2026)

## Abstract (arXiv)

As numerous instruction-tuning datasets continue to emerge, dynamically balancing and optimizing their mixtures has become a critical challenge. To address this, we propose DynamixSFT, a dynamic and automated method for instruction-tuning dataset mixture optimization. We formulate the problem as a multi-armed bandit setup and introduce a Prior-scaled Boltzmann Exploration that softly anchors the updated sampling distribution to the original dataset proportions, thereby preserving the inherent diversity and coverage of the collection. Sampling probabilities are updated using a lightweight 1-Step Look-ahead Reward, reflecting how much the dataset contributes to improving the model's performance at its current state. We demonstrate that DynamixSFT effectively optimizes the Tulu-2-mixture and Tulu-3-mixture collections across 10 benchmarks, while introducing minimal computational overhead over naive sampling. Furthermore, we provide a comprehensive analysis and visualizations to offer deeper insights into the adaptive dynamics of our method.