We have 4 of 28 papers
We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.
Not the right person? Other researchers publish under this name.
A continual data unlearning framework that uses completed deletion transitions as directional references to regularize future updates and achieves a better balance between deletion persistence and generative utility than existing unlearning baselines as requests accumulate.
A target-level evaluation protocol is introduced that tracks whether each target is forgotten immediately, remains forgotten at the end of the sequence, or reappears during subsequent deletions, and finds that targets that later reappear exhibit sharper local denoising-loss geometry after deletion than targets that rem...
This work proposes a circuit-grounded framework that connects training-dynamics-based data valuation with mechanistic interpretability (MI), and conceptualizes data quality along three complementary utility axes, learnability, challenge, and alignment, establishing a principled white-box paradigm for interpretable data...
STEP-KTODER is proposed, a framework for code preference optimization that defines steps as module-level functions in decomposed multi-function programs and assigns binary correctness labels via automatically generated unit tests and shows that execution-based labels are essential.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
