GRACE:Gradient-guided Coreset Selection for LLM Unlearning
Machine Unlearning methods for Large Language Models typically assume pre-specified forget and retain sets. In realistic settings, however, requests may provide only a few examples of undesired behavior, requiring forget and retain sets to be inferred from heterogeneous corpora. We study this data-selection problem and propose GRACE , a gradient-guided coreset selection method that constructs both forget and retain sets for LLM unlearning. GRACE first computes a forget direction from seed examples that elicit the undesired behavior, then selects a compact forget coreset whose gradients approximate this direction using non-negative orthogonal matching pursuit. To preserve model utility, it selects retain examples after projecting out the forget direction and applying clustered orthogonal matching pursuit in the remaining gradient space. Across two target domains, two model families, and four unlearning algorithms, GRACE improves model utility while maintaining comparable forget quality, with particularly consistent gains over prior gradient-based selection methods.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
GRACE: A Dynamic Coreset Selection Framework for Large Language Model Optimization
Large Language Models (LLMs) have demonstrated remarkable capabilities in natural language understanding and generation. However, their immense number of parameters and complex transformer-based architectures result in s…
Natural Language UnderstandingLLM Unlearning Reveals a Stronger-Than-Expected Coreset Effect in Current Benchmarks
Large language model unlearning has become a critical challenge in ensuring safety and controlled model behavior by removing undesired data-model influences from the pretrained model while preserving general utility. Sig…
Large Language ModelMachine UnlearningUPCORE: Utility-Preserving Coreset Selection for Balanced Unlearning
User specifications or legal frameworks often require information to be removed from pretrained models, including large language models (LLMs). This requires deleting or "forgetting" a set of data points from an already-…
TAGCOS: Task-agnostic Gradient Clustered Coreset Selection for Instruction Tuning Data
Instruction tuning has achieved unprecedented success in NLP, turning large language models into versatile chatbots. However, the increasing variety and volume of instruction datasets demand significant computational res…
Efficient Coreset Selection via K-Nearest Neighbor Graphs
Coreset selection reduces the cost of model training by replacing a large training set with a small representative subset. Existing gradient-approximation coreset methods such as CRAIG and cluster-based variants can pres…