# DPI: Exploiting Parameter Heterogeneity for Interference-Free Fine-Tuning

arXiv: https://arxiv.org/abs/2601.17777 · HTML: https://mlsanity.com/paper/2601.17777/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 32 references that are in the map.
- First posted on arXiv 25 Jan 2026.

## Builds on (most cited first, 25 of 25)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks](https://mlsanity.com/paper/1803.03635/) (2018)
- [Gradient Surgery for Multi-Task Learning](https://mlsanity.com/paper/2001.06782/) (2020)
- [Experience Replay for Continual Learning](https://mlsanity.com/paper/1811.11682/) (2018)
- [GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks](https://mlsanity.com/paper/1711.02257/) (2017)
- [AdapterFusion: Non-Destructive Task Composition for Transfer Learning](https://mlsanity.com/paper/2005.00247/) (2020)
- [TableBench: A Comprehensive and Complex Benchmark for Table Question Answering](https://mlsanity.com/paper/2408.09174/) (2024)
- [Time-aware Multiway Adaptive Fusion Network for Temporal Knowledge Graph Question Answering](https://mlsanity.com/paper/2302.12529/) (2023)
- [Question Calibration and Multi-Hop Modeling for Temporal Question Answering](https://mlsanity.com/paper/2402.13188/) (2024)
- [DABERT: Dual Attention Enhanced BERT for Semantic Matching](https://mlsanity.com/paper/2210.03454/) (2022)
- [Dual Path Modeling for Semantic Matching by Perceiving Subtle Conflicts](https://mlsanity.com/paper/2302.12530/) (2023)
- [Improving Semantic Matching through Dependency-Enhanced Pre-trained Model with Adaptive Fusion](https://mlsanity.com/paper/2210.08471/) (2022)
- [Searching for Optimal Subword Tokenization in Cross-domain NER](https://mlsanity.com/paper/2206.03352/) (2022)
- [Robust Lottery Tickets for Pre-trained Language Models](https://mlsanity.com/paper/2211.03013/) (2022)
- [Who Stole Your Data? A Method for Detecting Unauthorized RAG Theft](https://mlsanity.com/paper/2510.07728/) (2025)
- [Progressive Mastery: Customized Curriculum Learning with Guided Prompting for Mathematical Reasoning](https://mlsanity.com/paper/2506.04065/) (2025)
- [Not All Parameters Are Created Equal: Smart Isolation Boosts Fine-Tuning Performance](https://mlsanity.com/paper/2508.21741/) (2025)
- [Unleashing Potential of Evidence in Knowledge-Intensive Dialogue Generation](https://mlsanity.com/paper/2309.08380/) (2023)
- [Resolving Word Vagueness with Scenario-guided Adapter for Natural Language Inference](https://mlsanity.com/paper/2405.12434/) (2024)
- [Structural Reward Model: Enhancing Interpretability, Efficiency, and Scalability in Reward Modeling](https://mlsanity.com/paper/2509.25361/) (2025)
- [HoPE: Hyperbolic Rotary Positional Encoding for Stable Long-Range Dependency Modeling in Large Language Models](https://mlsanity.com/paper/2509.05218/) (2025)
- [Comateformer: Combined Attention Transformer for Semantic Sentence Matching](https://mlsanity.com/paper/2412.07220/) (2024)
- [DeCoRL: Decoupling Reasoning Chains via Parallel Sub-Step Generation and Cascaded Reinforcement for Interpretable and Scalable RLHF](https://mlsanity.com/paper/2511.19097/) (2025)

## Built on by (most cited first, 5 of 5)

- [Why Supervised Fine-Tuning Fails to Learn: A Systematic Study of Incomplete Learning in Large Language Models](https://mlsanity.com/paper/2604.10079/) (2026)
- [Reason Only When Needed: Efficient Generative Reward Modeling via Model-Internal Uncertainty](https://mlsanity.com/paper/2604.10072/) (2026)
- [Parameter Importance is Not Static: Evolving Parameter Isolation for Supervised Fine-Tuning](https://mlsanity.com/paper/2604.14010/) (2026)
- [Efficient Task Adaptation in Large Language Models via Selective Parameter Optimization](https://mlsanity.com/paper/2604.17051/) (2026)
- [Multi-Granularity Reasoning for Natural Language Inference](https://mlsanity.com/paper/2606.05181/) (2026)

## Abstract (arXiv)

Supervised fine-tuning (SFT) is a crucial step for adapting large language models (LLMs) to downstream tasks. However, conflicting objectives across heterogeneous SFT tasks often induce the "seesaw effect": optimizing for one task may degrade performance on others, particularly when model parameters are updated indiscriminately. In this paper, we propose a principled approach to disentangle and isolate task-specific parameter regions, motivated by the hypothesis that parameter heterogeneity underlies cross-task interference. Specifically, we first independently fine-tune LLMs on diverse SFT tasks and identify each task's core parameter region as the subset of parameters exhibiting the largest updates. Tasks with highly overlapping core parameter regions are merged for joint training, while disjoint tasks are organized into different stages. During multi-stage SFT, core parameters acquired in prior tasks are frozen, thereby preventing overwriting by subsequent tasks. To verify the effectiveness of our method, we conducted intensive experiments on multiple public datasets. The results showed that our dynamic parameter isolation strategy consistently reduced data conflicts and achieved consistent performance improvements compared to multi-stage and multi-task tuning baselines.