AI Paper Atlas

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

Mechanistic Finetuning of Vision-Language-Action Models via Few-Shot Demonstrations

Chancharik Mitra, Yusen Luo, Raj Saravanan, Dantong Niu, Anirudh Pai, Jesse Thomason, Trevor Darrell, Abrar Anwar, Deva Ramanan, Roei Herzig

ECCV 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 15 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
select task-specific heads via k-NN and finetune only them instead of full-head LoRApi0 / pi0.5 · 5 real-robot taskssuccess rate: e.g. Press Button Hard 45% -> 75% (pi0)Improved
finetune selected heads vs full LoRA, Press Button Hard on pi0.5pi0.5 · Press Button Hardsuccess rate: 90% -> 85%Hurt
use Robotic Steering instead of full-head LoRA for costpi0 · real-robot trainingtraining time / trainable params: 239 min -> 189 min; 1785.9M -> 78.8MImproved
single-task Robotic Steering under environment variationpi0.5 · Place Marker in Mug perturbationssuccess rate: lighting 25% -> 47.5%; form 22.5% -> 37.5%; distractor 30% -> 40%Improved
Robotic Steering for generalization to unseen Pick Mugpi0.5 · Pick Mug (unseen)success rate: 42.5% -> 65%Improved
multi-task: non-overlapping head selection vs joint LoRA (ablation)pi0.5 · Place Cube in Bowlsuccess rate: 37.5% -> 65%Improved
use k-NN regression instead of CMA for head selection (ablation)Robotic Steering · Place Marker in Mugsuccess rate: 15% -> 80%Improved
use k-NN regression instead of REINFORCE for head selection (ablation)Robotic Steering · Place Marker in Mughead selection time: success 80% vs 80%; time 93 min -> 0.2 minNo clear effect
also finetune MLPs of selected blocks (ablation)Robotic Steering · Place Marker in Mugsuccess rate: 10% -> 80%Improved
select top 20 heads vs all 144 heads (ablation)Robotic Steering · Place Marker in Mugsuccess rate: 45% -> 72.5%Improved
apply Robotic Steering to full finetuningpi0.5 · Place Marker in Mug / Press Button / Place Cubesuccess rate: 45% -> 60%; 77.5% -> 87.5%; 55% -> 72.5%Improved
select first token in suffix instead of last token in prefix (ablation)k-NN head selection · Place Marker in Mugcoefficient of variation: pi0.5: 0.05 -> 0.47; pi0: 0.04 -> 0.12Improved

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