Shadow-FT: Tuning Instruct Model via Training on Paired Base Model

• URL: http://arxiv.org/abs/2505.12716v3
• Date: Fri, 26 Sep 2025 03:43:47 GMT
• Title: Shadow-FT: Tuning Instruct Model via Training on Paired Base Model
• Authors: Taiqiang Wu, Runming Yang, Jiayi Li, Pengfei Hu, Yik-Chung Wu, Ngai Wong, Yujiu Yang,
• Abstract summary: Large language models (LLMs) consistently benefit from further fine-tuning on various tasks.<n>We propose a novel Shadow-FT framework to tune the Instruct models by leveraging the corresponding Base models.<n>Our proposed Shadow-FT introduces no additional parameters, is easy to implement, and significantly improves performance.
• Score: 67.20706292627106
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Large language models (LLMs) consistently benefit from further fine-tuning on various tasks. However, we observe that directly tuning the Instruct (i.e., instruction-tuned) models often leads to marginal improvements and even performance degeneration. Notably, paired Base models, the foundation for these Instruct variants, contain highly similar weight values (i.e., less than 2% on average for Llama 3.1 8B). The Base model tends to be a good learner yet a weak backbone without post-training. Therefore, we propose a novel Shadow-FT framework to tune the Instruct models by leveraging the corresponding Base models. The key insight is to fine-tune the Base model, and then \textit{directly} graft the learned weight updates to the Instruct model. Our proposed Shadow-FT introduces no additional parameters, is easy to implement, and significantly improves performance. We conduct extensive experiments on tuning mainstream LLMs, such as Qwen 3 and Llama 3 series, and evaluate them across 19 benchmarks covering coding, reasoning, and mathematical tasks. Experimental results demonstrate that Shadow-FT consistently outperforms conventional full-parameter and parameter-efficient tuning approaches. Further analyses indicate that Shadow-FT can be applied to multimodal large language models (MLLMs) and combined with direct preference optimization~(DPO). Codes and weights are available at \href{https://github.com/wutaiqiang/Shadow-FT}{Github}.

Related papers

• Do Instruction-Tuned Models Always Perform Better Than Base Models? Evidence from Math and Domain-Shifted Benchmarks [0.6536121591910934]
  We evaluate base and instruction-tuned models on standard math benchmarks, structurally perturbed variants, and domain-shifted tasks.<n>Our results show that base models surpass instruction-tuned variants on the domain-specific MedCalc benchmark.<n>In instruction-tuned models show sharp declines on perturbed datasets, indicating sensitivity to prompt structure over robust reasoning.
  arXiv  Detail & Related papers  (2026-01-19T17:26:49Z)
• Aligning Frozen LLMs by Reinforcement Learning: An Iterative Reweight-then-Optimize Approach [65.6966065843227]
  Iterative Reweight-then-IRO is a framework that performs RL-style alignment of a frozen base model without touching its parameters.<n>At test time, the value functions are used to guide the base model generation via a search-based optimization process.<n> Notably, users can apply IRO to align a model on their own dataset, similar to OpenAI's reinforcement fine-tuning (RFT)
  arXiv  Detail & Related papers  (2025-06-21T21:49:02Z)
• RoSTE: An Efficient Quantization-Aware Supervised Fine-Tuning Approach for Large Language Models [53.571195477043496]
  We propose an algorithm named Rotated Straight-Through-Estimator (RoSTE)<n>RoSTE combines quantization-aware supervised fine-tuning (QA-SFT) with an adaptive rotation strategy to reduce activation outliers.<n>Our findings reveal that the prediction error is directly proportional to the quantization error of the converged weights, which can be effectively managed through an optimized rotation configuration.
  arXiv  Detail & Related papers  (2025-02-13T06:44:33Z)
• SPaR: Self-Play with Tree-Search Refinement to Improve Instruction-Following in Large Language Models [88.29990536278167]
  We introduce SPaR, a self-play framework integrating tree-search self-refinement to yield valid and comparable preference pairs.<n>Our experiments show that a LLaMA3-8B model, trained over three iterations guided by SPaR, surpasses GPT-4-Turbo on the IFEval benchmark without losing general capabilities.
  arXiv  Detail & Related papers  (2024-12-16T09:47:43Z)
• 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. In-Context Learning (ICL) and. Efficient Fine-Tuning (PEFT) are currently two mainstream methods for augmenting. LLMs to downstream tasks. 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)
• ASFT: Aligned Supervised Fine-Tuning through Absolute Likelihood [14.512464277772194]
  Aligned Supervised Fine-Tuning (ASFT) is an effective approach that better aligns Large Language Models with pair-wise datasets. ASFT mitigates the issue where the DPO loss function decreases the probability of generating human-dispreferred data. Extensive experiments demonstrate that ASFT is an effective alignment approach, consistently outperforming existing methods.
  arXiv  Detail & Related papers  (2024-09-14T11:39:13Z)
• Comparative Analysis of Different Efficient Fine Tuning Methods of Large Language Models (LLMs) in Low-Resource Setting [0.0]
  We try to push the understanding of different fine-tuning strategies for large language models (LLMs) We compare state-of-the-art methods like vanilla fine-tuning and Pattern-Based Fine-Tuning (PBFT) on pre-trained models across two datasets, COLA and MNLI. Our findings suggest that these alternative strategies can exhibit out-of-domain generalization comparable to that of vanilla FT and PBFT.
  arXiv  Detail & Related papers  (2024-05-21T20:08:52Z)
• Insights into Alignment: Evaluating DPO and its Variants Across Multiple Tasks [35.36615140853107]
  This study evaluates Direct Preference Optimization (DPO) and its variants for aligning Large Language Models (LLMs) with human preferences.<n>Our evaluation spans 13 benchmarks covering dialogue, reasoning, mathematical problem-solving, question answering, truthfulness, MT-Bench, Big Bench, and the Open LLM Leaderboard.<n>We find that alignment methods often achieve near optimal performance even with smaller subsets of training data.
  arXiv  Detail & Related papers  (2024-04-23T03:55:01Z)
• Mixture-of-Experts Meets Instruction Tuning:A Winning Combination for Large Language Models [125.91897197446379]
  We find that MoE models benefit more from instruction tuning than dense models. Our most powerful model, FLAN-MOE-32B, surpasses the performance of FLAN-PALM-62B on four benchmark tasks.
  arXiv  Detail & Related papers  (2023-05-24T04:22:26Z)
• Re-parameterizing Your Optimizers rather than Architectures [119.08740698936633]
  We propose a novel paradigm of incorporating model-specific prior knowledge into Structurals and using them to train generic (simple) models. As an implementation, we propose a novel methodology to add prior knowledge by modifying the gradients according to a set of model-specific hyper- parameters. For a simple model trained with a Repr, we focus on a VGG-style plain model and showcase that such a simple model trained with a Repr, which is referred to as Rep-VGG, performs on par with the recent well-designed models.
  arXiv  Detail & Related papers  (2022-05-30T16:55:59Z)
• DSEE: Dually Sparsity-embedded Efficient Tuning of Pre-trained Language Models [152.29364079385635]
  As pre-trained models grow bigger, the fine-tuning process can be time-consuming and computationally expensive. We propose a framework for resource- and parameter-efficient fine-tuning by leveraging the sparsity prior in both weight updates and the final model weights. Our proposed framework, dubbed Dually Sparsity-Embedded Efficient Tuning (DSEE), aims to achieve two key objectives: (i) parameter efficient fine-tuning and (ii) resource-efficient inference.
  arXiv  Detail & Related papers  (2021-10-30T03:29:47Z)

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.