AI Paper Atlas

Research map / Reinforcement learning / Imitation learning

Making Learner Weakness Actionable for Learning from Demonstration with Novice Teachers

Yuqing Zhu, Matthew Howard

ICML 2026 · Reinforcement learning › Imitation learning · approach: Imitation combined with RL

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 CLASP-Teach demonstration guidance and difficulty-aware reweighting (CLASP-Full)diffusion policy · Robust success, 4 tasks at B=20success rate under shift: Avoiding 0.62 -> 0.88; Pushing 0.34 -> 0.72; P&P 0.44 -> 0.78; Alignment 0.16 -> 0.54 (vs no guidance)Improved
add difficulty-aware reweighting (CLASP-Train) on top of CLASP-Teach (ablation)CLASP · Alignment, B=20success rate: 0.56 -> 0.66Improved
add difficulty-aware reweighting (CLASP-Train) on top of CLASP-Teach (ablation)CLASP · Avoiding, B=20success rate: 0.92 -> 0.94No clear effect
add difficulty-aware reweighting (CLASP-Train) on top of CLASP-Teach (ablation)CLASP · Alignment, robust success under shift, B=20success rate: 0.42 -> 0.54Improved
use CLASP guidance instead of uncertainty-gated querying (ASkDAgger)diffusion policy · Robust success, Alignment B=20success rate under shift: 0.22 -> 0.54Improved
offline curation (quality or cluster weighting) without new guided demonstrationsno-guidance dataset · Robust success, Avoiding B=20success rate under shift: 0.62 -> 0.68 (quality), 0.70 (cluster); Alignment 0.16 -> 0.20, 0.18Mixed
easy-to-hard curriculum for demonstration collectiondiffusion policy · Robust success, Pushing B=20success rate under shift: 0.34 -> 0.58Improved
CLASP guidance on real robotdiffusion policy · Real-robot Avoiding, B=20success rate: 0.76 -> 0.92Improved
CLASP guidance on real robotdiffusion policy · Real-robot Stirring, B=20success rate: 0.68 -> 0.76Improved
CLASP guidance at very small budget on real robotdiffusion policy · Real-robot Avoiding, B=4success rate: 0.38 -> 0.32Hurt
CLASP guidance on completion time for easy taskdiffusion policy · Avoiding simulation, B=20normalised completion time: 0.34 -> 0.32No clear effect
add CLASP guidance for behavioural diversitydiffusion policy · Avoiding and Pushing strategy balancebalance score rImproved

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