# Towards Knowledge Alignment in Code LLMs: Contrastive Unlearning for Evolving APIs

arXiv: https://arxiv.org/abs/2606.30810 · HTML: https://mlsanity.com/paper/2606.30810/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Machine unlearning, one of 12 directions in LLM safety and security.
- Builds on 26 of its 32 references that are in the map.
- First posted on arXiv 29 Jun 2026.

## Placed in

- LLM safety and security / [Machine unlearning](https://mlsanity.com/safety/machine-unlearning/)

## Builds on (most cited first, 26 of 26)

- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence](https://mlsanity.com/paper/2401.14196/) (2024)
- [A Survey on Large Language Models for Code Generation](https://mlsanity.com/paper/2406.00515/) (2024)
- [The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning](https://mlsanity.com/paper/2403.03218/) (2024)
- [Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning](https://mlsanity.com/paper/2404.05868/) (2024)
- [StarCoder 2 and The Stack v2: The Next Generation](https://mlsanity.com/paper/2402.19173/) (2024)
- [Knowledge Unlearning for Mitigating Privacy Risks in Language Models](https://mlsanity.com/paper/2210.01504/) (2022)
- [RepairAgent: An Autonomous, LLM-Based Agent for Program Repair](https://mlsanity.com/paper/2403.17134/) (2024)
- [Large Language Models for Software Engineering: Survey and Open Problems](https://mlsanity.com/paper/2310.03533/) (2023)
- [Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearning](https://mlsanity.com/paper/2410.07163/) (2024)
- [CodeGen2: Lessons for Training LLMs on Programming and Natural Languages](https://mlsanity.com/paper/2305.02309/) (2023)
- [Can LLMs Replace Human Evaluators? An Empirical Study of LLM-as-a-Judge in Software Engineering](https://mlsanity.com/paper/2502.06193/) (2025)
- [On the Evaluation of Large Language Models in Unit Test Generation](https://mlsanity.com/paper/2406.18181/) (2024)
- [LLMs Meet Library Evolution: Evaluating Deprecated API Usage in LLM-based Code Completion](https://mlsanity.com/paper/2406.09834/) (2024)
- [HumanEvo: An Evolution-aware Benchmark for More Realistic Evaluation of Repository-level Code Generation](https://mlsanity.com/paper/2406.06918/) (2024)
- [Is Your LLM Outdated? A Deep Look at Temporal Generalization](https://mlsanity.com/paper/2405.08460/) (2024)
- [Mutation-Guided LLM-based Test Generation at Meta](https://mlsanity.com/paper/2501.12862/) (2025)
- [LibEvolutionEval: A Benchmark and Study for Version-Specific Code Generation](https://mlsanity.com/paper/2412.04478/) (2024)
- [ReCode: Updating Code API Knowledge with Reinforcement Learning](https://mlsanity.com/paper/2506.20495/) (2025)
- [Model Editing for LLMs4Code: How Far are We?](https://mlsanity.com/paper/2411.06638/) (2024)
- [Scrub It Out! Erasing Sensitive Memorization in Code Language Models via Machine Unlearning](https://mlsanity.com/paper/2509.13755/) (2025)
- [Large Language Model Unlearning for Source Code](https://mlsanity.com/paper/2506.17125/) (2025)
- [An Empirical Study on the Impact of Refactoring Activities on Evolving Client-Used APIs](https://mlsanity.com/paper/1709.09474/) (2017)
- [Don't Use a Cannon to Kill a Fly: Lightweight Model Editing for LLMs to Correct Deprecated API Recommendations](https://mlsanity.com/paper/2511.21022/) (2025)

## Built on by (most cited first, 1 of 1)

- [What Was Once Learned May Need to Be Unlearned: Machine Unlearning for Deprecated API Knowledge in Large Language Models](https://mlsanity.com/paper/2609.25786/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) have recently achieved strong performance in code generation. However, due to knowledge cut-off and the rapid evolution of software libraries, they often generate deprecated API usages that lead to unreliable and incompatible code. Existing fine-tuning methods lack selectivity when only a small portion of model knowledge requires modification. Recent model-level approaches, such as machine unlearning and model editing, offer a promising direction for modifying parametric knowledge. However, their use for deprecated API mitigation remains largely unexplored. Moreover, existing methods primarily suppress outdated APIs, but do not explicitly steer models toward correct replacements, often leading to mismatched or incomplete generations. To address this limitation, we developed CURE, a contrastive unlearning approach that shifts unlearning from purely suppressing outdated knowledge to explicitly promoting correct API replacements. Concretely, CURE jointly discourages deprecated APIs while encouraging their valid alternatives, enabling more reliable adaptation to evolving software libraries. The experiments on recent deprecated API benchmark dataset show that CURE not only reduces deprecated API usage but also improves correct API replacement, while preserving general code generation performance. CURE substantially outperforms two SOTA baselines with respect to different quality metrics. These findings highlight the importance of combining suppression with replacement when adapting LLMs to evolving software ecosystems.