AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Probing the Knowledge Boundary: An Interactive Agentic Framework for Deep Knowledge Extraction

Yuheng Yang, Siqi Zhu, Tao Feng, Ge Liu, Jiaxuan You

ICML 2026 · Large language models › Retrieval and knowledge · approach: Knowledge editing in LLMs · 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
use Recursive Taxonomy L5W5 instead of Taxonomy L2W2 (ablation)Llama-3.1-405B probing · Strategy comparisonyield ratio: 1.00 -> 5.98Improved
increase taxonomy width/depth from L2W2 to L3W3 (ablation)Recursive Taxonomy · Strategy comparisonyield ratio: 1.00 -> 2.15Improved
replace Sequential probing with Recursive Taxonomy L5W5Llama-3.1-405B probing · Strategy comparisonyield ratio: 0.28 -> 5.98Improved
add self-reflection critic loop to sequential probing (ablation)Sequential probing · Strategy comparisonyield ratio: 0.28 -> 0.32Improved
increase expert personas from N=3 to N=20 (ablation)Multi-Perspective probing · Strategy comparisonyield ratio: 1.42 -> 4.19Improved
scale model from Llama-3.1-8B to 70B (ablation)Taxonomy-L5W5 · Scaling, average of 3 domainsrecall: 27.3% -> 36.9%Improved
scale model from 70B to 405B (ablation)Taxonomy-L5W5 · Scaling, average of 3 domainsrecall: 36.9% -> 39.0%Improved
scale model from 70B to 405B (ablation)Taxonomy-L5W5 · Scaling, average of 3 domainsaccuracy: 89.0% -> 87.7%Hurt
scale model from 8B to 70B (ablation)Taxonomy-L5W5 · Scaling, average of 3 domainsaccuracy: 81.2% -> 89.0%Improved
swap Qwen2.5-7B Instruct for RL-tuned Coder (ablation)Qwen2.5-7B · Specialization, Probabilistic Methodsfinal recall: 66.7% -> 34.0%Hurt
swap Qwen2.5-7B Instruct for RL-tuned Coder (ablation)Qwen2.5-7B · Specialization, Deep Learningfinal recall: 57.7% -> 42.2%Hurt
swap Qwen2.5-7B Instruct for RL-tuned Coder (ablation)Qwen2.5-7B · Specialization, Deep Learninginitial accuracy: 92.3% -> 96.4%Improved

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