AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Mitigating Error Accumulation in Knowledge Editing for Multi-Hop Question Answering

Jiaxin Guo, Hao Sun, Wenhao Zhang, Xuanbo Fan, Yan Zhang

AAAI 2026 · Large language models › Retrieval and knowledge · approach: Knowledge editing in LLMs

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
replace tree/chain-based editing with tree-structured evidence expansion (ToE)ToE · MQuAKE-CF reliability (Llama-3-8B)accuracy: 48.3 (RAE-prune) -> 53.1Improved
use NLI observer instead of Llama-3-8B LLM as conflict detector (ablation)ToE · MQuAKE-CF reliabilityaccuracy: 38.1 -> 53.1Improved
use NLI observer instead of GPT-4o-mini as conflict detector (ablation)ToE · MQuAKE-CF reliabilityaccuracy: 51.8 -> 53.1Improved
use NLI observer instead of Qwen2.5-32B LLM observer (ablation)ToE · MQuAKE-CF reliabilityaccuracy: 49.9 -> 53.1Improved
expand top-b scoring nodes rather than bottom-b (ablation)ToE · MQuAKE-CF reliabilityaccuracy: 23.4 -> 53.1Improved
expand top-b scoring nodes rather than bottom-b (ablation)ToE · DUNE New Infoaccuracy: 81.0 -> 93.3Improved
increase beam width and branching from b=1,n=1 to b=2,n=3 (ablation)ToE · MQuAKE-CF reliabilityaccuracy: 42.0 -> 53.1Improved
increase search further from b=2,n=3 to b=3,n=5 (ablation)ToE · MQuAKE-CF reliabilityaccuracy: 53.1 -> 52.7 (AVG 76.2 -> 77.0)No clear effect
apply ToE to GPT-4o-mini and GPT-4o versus RAEToE · MQuAKE-CF reliability (black-box LLMs)accuracy: GPT-4o-mini 55.3 (RAE) vs 54.1; GPT-4o 58.7 (RAE) vs 60.8Mixed
use multi-GPU parallel branch explorationToE · MQuAKE-CF latency (300 examples)latencyImproved
tree expansion versus chain-based reasoning on single GPUToE · MQuAKE-CF latencylatency: 2-3x higher latencyHurt
ToE versus EditCoT on DUNE ScienceToE · DUNE Science (Llama-3-8B)accuracy: 85.0 (EditCoT) -> 82.2Hurt

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