AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

ActiShade: Activating Overshadowed Knowledge to Guide Multi-Hop Reasoning in Large Language Models

Huipeng Ma, Luan Zhang, Dandan Song, Linmei Hu, Yuhang Tian, Jun Yang, Changzhi Zhou, Chenhao Li, Yizhou Jin, Xudong Li, Meng Lin, Mingxing Zhang and 1 more

AAAI 2026 · Large language models › Retrieval and knowledge · approach: RAG robustness and evaluation · 0 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 Knowledge Overshadowing Detection (GaP) to multi-round ActiShade (ablation)ActiShade (Llama-3-8B) · MuSiQueF1: 22.83 -> 26.94Improved
add GaP detection to ActiShade (ablation)ActiShade (Llama-3-8B) · HotpotQAF1: 51.23 -> 56.33Improved
add GaP detection to ActiShade (ablation)ActiShade (Llama-3-8B) · 2WikiMQAF1: 45.18 -> 46.02Improved
add GaP detection to single-round retrieval (ablation)Direct-R (Llama-3-8B) · 2WikiMQAF1: 35.56 -> 38.67Improved
replace GaP by CoDA token-removal detection (ablation)ActiShade (Llama-3-8B) · 2WikiMQAF1: 46.02 -> 41.29Hurt
add CoDA detection to ActiShade without detection (ablation)ActiShade-NoKOD (Llama-3-8B) · MuSiQueF1: 22.83 -> 21.23Hurt
add CoDA detection to single-round retrieval (ablation)Direct-R (Llama-3-8B) · HotpotQAF1: 44.89 -> 45.98Improved
train retriever with fine-grained contrastive loss (positive/semi-positive/negative) instead of standard contrastive (ablation)ActiShade retriever · MuSiQueF1: 24.10 -> 26.94Improved
add fine-grained contrastive retriever fine-tuning vs base retriever (ablation)contriever-msmarco · Retriever test, positive docsRecall@1: 29.20 -> 75.33Improved
fine-grained instead of standard contrastive loss for positive plus semi-positive docs at Recall@1 (ablation)ActiShade retriever · Retriever test, Pos&SemiRecall@1: 38.21 -> 38.14No clear effect
vary Gaussian noise sigma from 0.05 to 0.5 in GaP (ablation)ActiShade (Llama-3-8B) · multi-hop QAF1Mixed
scale generator from Qwen2.5-7B to 14BActiShade · HotpotQAF1: 55.45 -> 57.45Improved

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