Search papers, labs, and topics across Lattice.
This paper introduces Hybrid-Policy Self-Editing (HPSE), a novel approach to knowledge editing in large language models (LLMs) that enhances composability of edited knowledge. By proactively distilling knowledge from a privileged in-context state of the model, HPSE addresses the limitations of existing unstructured knowledge editing methods, which fail to enable atomic question answering and multi-hop reasoning. Empirical evaluations demonstrate that HPSE significantly improves the performance of multiple LLM backbones and knowledge editors, highlighting its effectiveness in integrating new information while maintaining model integrity.
HPSE transforms how LLMs integrate new knowledge, enabling them to answer atomic questions and perform multi-hop reasoning with edited facts.
Large language models (LLMs) achieve remarkable performance across natural language tasks, yet they are trained on static corpora and their knowledge quickly becomes outdated in a fast-changing world. This motivates knowledge editing (KE), which updates specific knowledge in an LLM without changing unrelated others. Recent works move from structured knowledge triples toward unstructured KE (UKE), where the edit is a free-form passage that may state multiple facts at once. Nonetheless, existing editors inject such a passage yet fail to use it: the edited model can recall the passage, but can neither answer atomic questions about its facts nor compose them into multi-hop reasoning. We attribute this missing property, which we term composability, to editors'passive reliance on the fixed passage as the sole learning source. In response, we cast editing as a proactive self-distillation from a privileged in-context state of the same model, which requires no external supervision. We further reveal that due to the novelty of the injected knowledge, the pre-edited model's own rollouts rarely cover it, which limits the effectiveness of pure on-policy distillation. To close this gap, we propose HPSE, which builds a hybrid rollout that steps in to place missing facts onto the student's own trajectory precisely where its coverage fails, while staying on-policy elsewhere. We theoretically analyze HPSE's advantage over pure on-policy distillation, and empirically establish its plug-and-play improvements across four LLM backbones and two KE editors under various scenarios.
