AI Paper Atlas

Research map / Multi-agent systems and games / LLM agents

Agentic Model Predictive Questioning Control in Visual Design

Kuang-Da Wang, Zhao Wang, Wei-Yao Wang, Yotaro Shimose, Jaechang Kim, Shingo Takamatsu

ICML 2026 · Multi-agent systems and games › LLM agents · approach: Agentic benchmarks and orchestration

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes Agentic Model Predictive Questioning Control (A-MPQC), a test-time framework for LLM-based visual design clarification that accounts for the user's cognitive burden when choosing questions.

Problem

LLM-based approaches to clarifying visual design intent focus on selecting informative questions but overlook the cognitive burden of interpreting and answering them, which matters for human-agent interaction.

Approach

A test-time, model-predictive control style questioning policy (A-MPQC) that plans which clarifying questions to ask while explicitly modeling the user's cognitive effort in answering them, rather than only maximizing intent uncovering.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
remove question plans (lookahead) from A-MPQC (ablation)A-MPQC · MIMOΔS: 0.594 -> 0.391 when removedImproved
add respond-or-reject surrogate reward instead of always-answer (ablation)A-MPQC · MIMOΔS: 0.204 -> 0.594Improved
use rejection feedback in replanning rather than ignoring it (ablation)A-MPQC · MIMOΔS/C efficiency: 0.361 -> 0.480Improved
add adaptive question format control instead of direct generation (ablation)A-MPQC · MIMOΔS/C efficiency: 0.155 -> 0.480Improved
apply MPC-style control over within-round question dimension instead of rounds (ablation)A-MPQC · MIMOinteraction cost: 1.238 -> 6.696Hurt
add format flexibility (binary/MC/open) to direct generationDG · MIMOΔS/C efficiency: 0.038 -> 0.197Improved
add retrieval augmentation to DG + FlexibleDG + Flexible · WebGen-V and MIMOΔS: WebGen-V 0.418 -> 0.383; MIMO 0.457 -> 0.292Mixed
split budget B=12 into intermediate n x m allocation (3x4) rather than extremes (ablation)A-MPQC · budget allocationΔSImproved
use fixed binary-only question formatDG · MIMOΔS: 0.019 vs 0.256 for DGHurt

Experimental setup

Builds on: model predictive control

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