AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

Describe-Then-Act: Proactive Agent Steering via Distilled Language-Action World Models

Massimiliano Pappa, Luca Romani, Valentino Sacco, Alessio Palma, Stéphane Lathuilière, Fabio Galasso, Xavier Alameda-Pineda, Indro Spinelli

ECCV 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 1 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 the three-stage curriculum (projector alignment, description distillation) before verdict training (ablation)DILLO-1B · LIBERO avgsuccess rate (%): 71.6 -> 78.0Improved
add the three-stage curriculum (ablation)DILLO-4B · LIBERO avgsuccess rate (%): 73.5 -> 78.9Improved
add the three-stage curriculum (ablation)DILLO-4B · LIBERO-Spatialsuccess rate (%): 54.0 -> 73.0Improved
add description generation objective to verdict training (ablation)DILLO-4B · LIBERO avgsuccess rate (%): 72.2 -> 78.9Improved
add description generation objective to verdict training (ablation)DILLO-1B · LIBERO avgsuccess rate (%): 75.1 -> 78.0Improved
use language-model DILLO-4B instead of latent MLP critic (ablation)DILLO-4B · LIBERO avgsuccess rate (%): 74.2 -> 78.9Improved
steer base policy with latent MLP critic (ablation)Base policy · LIBERO avgsuccess rate (%): 74.7 -> 74.2No clear effect
steer with DILLO-4B rejection sampling (N=5)SAC policy · MetaWorld Soccersuccess rate (%): 65.0 -> 80.0Improved
steer with DILLO-1B rejection samplingACT policy · LIBERO-90success rate (%): 71.4 -> 68.6Hurt
use heavier rejection gating threshold on LIBERO-90 (ablation)DILLO · LIBERO-90success rate: -4.4%Hurt
steer LIBERO-10 policy with LIBERO-90-trained DILLO (cross-suite)ACT policy · LIBERO-10success rate (%): 64.0 -> 69.5Improved
scale student from 1B to 4B (ablation)DILLO · MetaWorld steeringsuccess rate (%): Drawer-Open 70.0 -> 90.0; Sweep-Into 90.0 -> 85.0Mixed

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