# Unlocking the Power of Function Vectors for Characterizing and Mitigating Catastrophic Forgetting in Continual Instruction Tuning

arXiv: https://arxiv.org/abs/2502.11019 · HTML: https://mlsanity.com/paper/2502.11019/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- Builds on 54 of its 74 references that are in the map.
- First posted on arXiv 16 Feb 2025.

## Builds on (most cited first, 30 of 54)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [Scaling Instruction-Finetuned Language Models](https://mlsanity.com/paper/2210.11416/) (2022)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answering](https://mlsanity.com/paper/1809.02789/) (2018)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models](https://mlsanity.com/paper/2206.04615/) (2022)
- [Gradient Episodic Memory for Continual Learning](https://mlsanity.com/paper/1706.08840/) (2017)
- [CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge](https://mlsanity.com/paper/1811.00937/) (2018)
- [Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small](https://mlsanity.com/paper/2211.00593/) (2022)
- [Efficient Lifelong Learning with A-GEM](https://mlsanity.com/paper/1812.00420/) (2018)
- [Mass-Editing Memory in a Transformer](https://mlsanity.com/paper/2210.07229/) (2022)
- [The LAMBADA dataset: Word prediction requiring a broad discourse context](https://mlsanity.com/paper/1606.06031/) (2016)
- [An Explanation of In-context Learning as Implicit Bayesian Inference](https://mlsanity.com/paper/2111.02080/) (2021)
- [Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Tasks](https://mlsanity.com/paper/2204.07705/) (2022)
- [Instruction Tuning with GPT-4](https://mlsanity.com/paper/2304.03277/) (2023)
- [Overcoming catastrophic forgetting with hard attention to the task](https://mlsanity.com/paper/1801.01423/) (2018)
- [An Empirical Study of Catastrophic Forgetting in Large Language Models During Continual Fine-tuning](https://mlsanity.com/paper/2308.08747/) (2023)
- [Orthogonal Subspace Learning for Language Model Continual Learning](https://mlsanity.com/paper/2310.14152/) (2023)
- [In-Context Learning Creates Task Vectors](https://mlsanity.com/paper/2310.15916/) (2023)
- [Function Vectors in Large Language Models](https://mlsanity.com/paper/2310.15213/) (2023)
- [How is ChatGPT's behavior changing over time?](https://mlsanity.com/paper/2307.09009/) (2023)

## Built on by (most cited first, 16 of 16)

- [Continual Learning for Generative AI: From LLMs to MLLMs and Beyond](https://mlsanity.com/paper/2506.13045/) (2025)
- [FOREVER: Forgetting Curve-Inspired Memory Replay for Language Model Continual Learning](https://mlsanity.com/paper/2601.03938/) (2026)
- [Decomposing the Basic Abilities of Large Language Models: Mitigating Cross-Task Interference in Multi-Task Instruct-Tuning](https://mlsanity.com/paper/2605.05676/) (2026)
- [Scalable Strategies for Continual Learning with Replay](https://mlsanity.com/paper/2505.12512/) (2025)
- [Task Vectors, Learned Not Extracted: Performance Gains and Mechanistic Insight](https://mlsanity.com/paper/2509.24169/) (2025)
- [Learning from the Undesirable: Robust Adaptation of Language Models without Forgetting](https://mlsanity.com/paper/2511.13052/) (2025)
- [Contrastive Regularization over LoRA for Multimodal Biomedical Image Incremental Learning](https://mlsanity.com/paper/2508.11673/) (2025)
- [DFAMS: Dynamic-flow guided Federated Alignment based Multi-prototype Search](https://mlsanity.com/paper/2508.20353/) (2025)
- [DORAEMON: A Unified Library for Visual Object Modeling and Representation Learning at Scale](https://mlsanity.com/paper/2511.04394/) (2025)
- [Grow, Don't Overwrite: Fine-tuning Without Forgetting](https://mlsanity.com/paper/2603.08647/) (2026)
- [Learn from Weaknesses: Automated Domain Specialization for Small Computer-Use Agents](https://mlsanity.com/paper/2605.28775/) (2026)
- [iDocV2: Leveraging Self-Supervision and Open-Set Detection for Improving Pattern Spotting in Historical Documents](https://mlsanity.com/paper/2604.16726/) (2026)
- [Low-Rank Adapters Initialization via Gradient Surgery for Continual Learning](https://mlsanity.com/paper/2605.12752/) (2026)
- [Training Prompt Matters: State-Adaptive Optimization for Robust Fine-Tuning](https://mlsanity.com/paper/2606.01967/) (2026)
- [Few-Shot Demonstrations Elicit the Use of In-Context World Representations in LLMs](https://mlsanity.com/paper/2609.24352/) (2026)
- [Towards Eliminating Catastrophic Forgetting in the Curriculum Learning of Math Reasoning Tasks](https://mlsanity.com/paper/2609.33655/) (2026)

## Abstract (arXiv)

Catastrophic forgetting (CF) poses a significant challenge in machine learning, where a model forgets previously learned information upon learning new tasks. Despite the advanced capabilities of Large Language Models (LLMs), they continue to face challenges with CF during continual learning. The majority of existing research focuses on analyzing forgetting patterns through a singular training sequence, thereby overlooking the intricate effects that diverse tasks have on model behavior. Our study explores CF across various settings, discovering that model forgetting is influenced by both the specific training tasks and the models themselves. To this end, we interpret forgetting by examining the function vector (FV), a compact representation of functions in LLMs, offering a model-dependent indicator for the occurrence of CF. Through theoretical and empirical analyses, we demonstrated that CF in LLMs primarily stems from biases in function activation rather than the overwriting of task processing functions. Leveraging these insights, we propose a novel function vector guided training methodology, incorporating a regularization technique to stabilize the FV and mitigate forgetting. Empirical tests on four benchmarks confirm the effectiveness of our proposed training method, substantiating our theoretical framework concerning CF and model function dynamics. We plan to make our code publicly accessible in the near future.