AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Escaping the Diversity Trap in Robotic Manipulation via Anchor-Centric Adaptation

Yanzhe Chen, Kevin Yuchen Ma, Qi Lv, Yiqi Lin, Zechen Bai, Chen Gao, Mike Zheng Shou

ICML 2026 · Reinforcement learning › Visual and representation RL · approach: Dexterous manipulation transfer · 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 Anchor-Centric Adaptation (repeated anchors + mined boundary + residual) to standard diverse fine-tuningπ0.5 · Real-robot 4 tasks, N=100mean success rate (%): 31.7 -> 72.5Improved
add ACA at N=50 versus baseline at N=150π0.5 · Real-robot 4 tasksmean success rate (%): 46.3 vs 52.9Improved
use error-driven boundary mining instead of random expansion (with full FT) (ablation)ACA · Real-robot, N=100mean success rate (%): 57.1 -> 64.2Improved
use residual (LoRA) update instead of full fine-tuning, with mined expansion (ablation)ACA · Real-robot, N=100mean success rate (%): 64.2 -> 72.5Improved
use residual update with random expansion instead of full FT (ablation)ACA · Real-robot, N=100mean success rate (%): 57.1 -> 58.8Improved
add Stage 2 with random expansion and full fine-tuning on top of Stage-1 policy (ablation)ACA · Real-robot, N=100mean success rate (%): 62.1 -> 57.1Hurt
increase number of distinct anchors beyond the optimum (ablation)ACA · Real-robot, budgets 50/100/150success rateHurt
reduce anchor budget NA from 80 to 50 (ablation)ACA · Real-robot, N=100S@1 success rate (%): 80.0 -> 65.0Hurt
raise NA from 80 to 100 (fewer Stage 2 trajectories) (ablation)ACA · Real-robot, N=100S@3 / mean success: S@1 82.5; S@3 55.0 vs 65.0Mixed
use random anchor placement instead of structured Center Rect layout (ablation)ACA · Real-robot, N=100, K=6mean success rate (%): 72.5 -> 70.4 / 68.8No clear effect
use spatially biased top-left anchors instead of no anchors (ablation)ACA · Real-robot, N=100S@3 success rate (%): 12.5 -> 51.0Improved
use rectangular vs circular anchor arrangement (ablation)ACA · Real-robot, N=100region success ratesNo clear effect

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