AI Paper Atlas

Research map / Large language models / Agents, code and math

JADE: Expert-Grounded Dynamic Evaluation for Open-Ended Professional Tasks

Lanbo Lin, Jiayao Liu, Tianyuan Yang, Li Cai, Yuanwu Xu, Lei Wei, Sicong Xie, Guannan Zhang

ICML 2026 · Large language models › Agents, code and math · approach: Inductive and abstract reasoning 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
Use full two-layer JADE instead of vanilla pointwise LLM-as-a-judgeJADE · BizBench human alignmentPearson r: 0.667 -> 0.858Improved
Add expert skill activation (Layer 1) (ablation)JADE · BizBench human alignmentPearson r: 0.769 -> 0.844Improved
Add report-specific checklist (Layer 2) (ablation)JADE · BizBench human alignmentPearson r: 0.769 -> 0.786Improved
Combine both layers rather than either alone (ablation)JADE · BizBench human alignmentPearson r: 0.844 -> 0.858Improved
Use a flat generated checklist instead of vanilla judge (ablation)LLM judge · BizBench human alignmentPearson r: 0.667 -> 0.769Improved
Swap JADE backbone from GPT-5-0807 to GPT-4.1 or Gemini-3-flash (ablation)JADE · BizBench human alignmentPearson r: 0.858 -> 0.827 (GPT-4.1); 0.858 -> 0.841 (Gemini-3-flash)Hurt
Add evidence verification agent to JADE (ablation)JADE · BizBench human alignmentPearson r: 0.844 -> 0.858Improved
Add domain skills summarized from rubrics (ablation)JADE · HealthBench subset (n=474)Spearman rho: 0.154 -> 0.228Improved
Add domain skills on HealthBench after trimming 20% of items with specific clinical details (ablation)JADE · HealthBench trimmed 20%Spearman rho: 0.187 -> 0.530Improved
Apply JADE with auto-summarized skills to sports domainJADE · DR.BENCH Sports & competitionsPearson r: 0.478 (vs 0.736 overall)Hurt
Enable tool use for agentsClaude Opus 4.5 · BizBench final scorefinal score (%): 32.1 -> 36.2Improved
Enable tool use for agentsGemini 3 Pro · BizBench final scorefinal score (%): 44.8 -> 41.8Hurt

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