AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Multi-granularity Temporal Knowledge Editing over Large Language Models

Simiao Zhao, Ning Pang, Zhen Tan, Yanli Hu, Weidong Xiao, Xiang Zhao

AAAI 2026 · Large language models › Retrieval and knowledge · approach: Knowledge editing in LLMs · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add time-anchored mechanism (timestamp anchors) (ablation)SPIKE · MTKE-SGACC current: 50.11 -> 84.37Improved
add time-anchored mechanism (timestamp anchors) (ablation)SPIKE · MTKE-MGACC current: 42.82 -> 75.69Improved
add time-anchored mechanism (ablation)SPIKE · MTKE-SGHTC: 0.6884 -> 0.7926Improved
add time-anchored mechanism (ablation)SPIKE · MTKE-MGHTC: 0.6640 -> 0.6904Improved
add time-anchored mechanism (ablation)SPIKE · MTKE-SGACC history: 82.66 -> 85.01Improved
use SPIKE instead of best METO-based baselineSPIKE (GPT-J) · MTKE-SGACC history: 26.55 -> 85.01Improved
use SPIKE instead of best METO-based baselineSPIKE (GPT-J) · MTKE-MGACC history: 29.83 -> 80.66Improved
use SPIKE instead of best METO-based baselineSPIKE (GPT-J) · MTKE-SGHTC: 0.5734 -> 0.7926Improved
use SPIKE instead of ROMESPIKE (GPT-J) · MTKE-SGACC current: 91.22 -> 84.37Hurt
use SPIKE instead of ROMESPIKE (GPT-J) · MTKE-MGACC current: 92.27 -> 75.69Hurt
use SPIKE instead of ROMESPIKE (GPT-J) · MTKE-SGSoft Recall: 0.4820 -> 0.3552Hurt
filter AToKe samples where model already knows target or lacks historySPIKE · AToKe-SEACC history: 20.25 -> 29.86Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy