AI Paper Atlas

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

Beyond Gemini-3-Pro: Revisiting LLM Routing and Aggregation at Scale

Shengji Tang, Weihao Lin, Peng Ye, Jingqi Ye, Hao Li, Yiqun Zhang, Xiaosong Wang, Bo Zhang, Shuyue Hu, Tao Chen, LEI BAI, Wanli Ouyang

ICML 2026 · Large language models › Agents, code and math · approach: LLMs for optimization and heuristics · 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 adaptive routing-aggregation switch (ARS) (ablation)JiSi (QRR+SAS) · Component ablation (average over three domains)average score: 71.09 -> 72.15Improved
add support-set-based aggregator selection (SAS) (ablation)JiSi (QRR) · Component ablation (average over three domains)average score: 69.68 -> 71.09Improved
add query-response mixed routing (QRR) (ablation)JiSi · Component ablation (average over three domains)average score: 67.83 -> 69.68Improved
add ARS alone without QRR and SAS (ablation)baseline aggregation · Component ablation (average over three domains)average score: 67.83 -> 64.87Hurt
add SAS alone (ablation)baseline aggregation · Component ablation (average over three domains)average score: 67.83 -> 68.74Improved
add response similarity to query similarity in support set (ablation)JiSi routing · Similarity criteriondeviation (lower better): 34.40 -> 32.96Improved
add length (token cost) similarity (ablation)JiSi routing · Similarity criteriondeviation (lower better): 32.96 -> 31.27Improved
apply adaptive switch to filter references on Hard questions (ablation)JiSi · Reference accuracy, Hardaccuracy: 22.25 -> 30.74Improved
apply adaptive switch on Easy questions (ablation)JiSi · Reference accuracy, Easyaccuracy: 95.37 -> 98.07Improved
aggregate with JiSi instead of routing only (query-response mixed router) (ablation)JiSi · Nine-benchmark average (LLMRouterBench 30% test split)average score: 69.68 -> 72.15Improved
scale majority voting from 3 best LLMs to 10 LLMsMajority Voting · Nine-benchmark average (LLMRouterBench 30% test split)average score: 64.64 -> 61.69Hurt
use four rollouts self-consistency instead of one best LLMBest LLM · Nine-benchmark average (LLMRouterBench 30% test split)average score: 70.55 -> 70.66No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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