AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Small Agent Group is the Future of Digital Health

Yuqiao Meng, Luoxi Tang, Dazheng Zhang, Rafael Brens, Elvys J. Romero, Nancy Guo, Safa elkefi, Zhaohan Xi

ICML 2026 · Large language models › Retrieval and knowledge · approach: Long-context and personalized retrieval · 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
replace single 70B/72B LLM with Small Agent Group of small agents (pre-trained)single giant LLM · MedQA, MedMCQA, NEJM-MedQA, GPQA-Bioaccuracy: MedQA 72.0 -> 86.0 (Qwen); 59.8 -> 84.6 (Llama)Improved
add GRPO optimization to SAGSAG (pre-trained) · GPQA-Bio, Qwen backboneaccuracy: 73.3 -> 92.6Improved
add CTDE optimization to SAGSAG (pre-trained) · NEJM-MedQA, Llama backboneaccuracy: 64.9 -> 86.7Improved
add reasoning agents (AR) (ablation)SAG · GPQA-Bio, Llama backboneaccuracy: 34.7 -> 70.1Improved
add knowledge agents with RAG (AK) (ablation)SAG · GPQA-Bio, Llama backboneaccuracy: 45.6 -> 70.1Improved
add judgment agents (AJ) (ablation)SAG · GPQA-Bio, Llama backboneaccuracy: 47.2 -> 70.1Improved
add safety agents (AS) (ablation)SAG · GPQA-Bio, Llama backboneaccuracy: 59.0 -> 70.1Improved
use multi-agent debate instead of majority voting (ablation)SAG · GPQA-Bio, Llama backboneaccuracy: 44.0 -> 70.1Improved
use SAG instead of single Qwen-72B for demographic fairnesssingle LLM · EquityMedQAavg CDR: 2.4% -> 1.4%Improved
use SAG under decoding variation instead of single giant LLMsingle giant LLM · NEJM-MedQA consistency (Cand. setting, Llama)std over 5 runs: ±5.47 -> ±2.12Improved
mix Llama and Qwen backbones in SAG instead of homogeneous groups (ablation)SAG · NEJM-MedQA, 10 agents, CTDEaccuracy: 85.6 -> 91.3 (vs Qwen-only)Improved
use SAG instead of single giant LLM for deploymentsingle 72B LLM · peak memory, latency, computememory / latency / TFLOPs: memory 79.5-96.8 GB vs 130 GB+; latency 58.7-63.9 s vs 27-28 s; 28.6-31.4 vs 23-25 TFLOPsMixed

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