Rho: A Foundation for Efficiently Adaptable VLA Models

Rho Team; Simran Bagaria; Daphne Chen; Dean Fortier; Jianlong Fu; Michael Harrison; Tess Hellebrekers; Neel Joshi

cs.ROSep 29, 2026

Rho: A Foundation for Efficiently Adaptable VLA Models

Authors: Rho Team, Simran Bagaria, Daphne Chen, Dean Fortier, Jianlong Fu, Michael Harrison, Tess Hellebrekers, Neel Joshi, +6 more

Organizations: Microsoft Research

Abstract

General-purpose physical AI models must combine broad visual and linguistic capabilities with precise control across robot embodiments and efficient adaptation to downstream tasks. We introduce Rho, a family of open-weights VLA models for bimanual manipulation designed for data-light task adaptation on 3 embodiments representative of dual-arm robots across research labs and the industry -- YAM Box, UR AI Trainer, and FR3 Duo. We systematically ablate Rho's action-expert architecture and training recipe, and show in controlled simulation and physical-robot experiments that embodiment midtraining improves downstream adaptation. The resulting Rho variants for YAM Box, UR AI Trainer, and FR3 Duo match or outperform existing open-weights VLAs and achieve the strongest overall performance across the tasks, embodiments, and baselines evaluated in this report. We further demonstrate the Rho model family's built-in capacity for online adaptation: an internal latent policy learns from corrective feedback to select observation-conditioned noise inputs for the frozen flow-matching action expert. With as few as 15 corrected episodes, adapting this lightweight module enables Rho to handle task situations at the fringe of its offline finetuning distribution. Together, these results position Rho as both a strong general-purpose robotic manipulation model and a practical foundation for adaptation. We release the base Rho model and the embodiment-specific checkpoints to facilitate Rho's deployment in research experiments and practical industrial use cases.

Figures & tables

Panel of Figure 1 — Figure 1 : Rho’s physical dual-arm robot embodiments. YAM Box is common in research labs, FR3 Duo’s usage straddles research and industry, and UR AI Trainer is an emerging standard setup for industrial bimanual manipulation tasks. Midtrained checkpoints for all three are available at https://huggingface.co/collections/microsoft/rho

Component	Specification
Full vision-language model	4.68B parameters
Language decoder	32 layers; width 3,072; MLP width 8,192
Decoder attention	32 query heads; 32 key/value heads
Vocabulary and positions	100,352 tokens; 16,384 configured positions
Vision encoder	SigLIP 2 SO400M NaFlex; 428M parameters
Image representation	$16\!\times\!16$ patches; 256–3,600 visual tokens

Table 1: Architecture of the Phi-Phy vision-language backbone used to initialize Rho.

Name	Description	Example
Attribute Similarity	Find the odd-one-out by font, color, or layout.	Which screenshot uses a different font?
Component Counting	Count a UI component across screenshots.	How many screenshots contain a card?
Object Counting	Count objects across natural images.	How many boats in all images?
Chart Type Matching	Find the odd-one-out by chart type.	Which chart is not a grouped bar chart?
Cross-Chart Reasoning	Combine evidence from related charts.	What patterns emerge across the charts?
Chart Difference Spotting	Identify chart changes and their effects.	What changed; how were profits affected?

Table 2: Multi-image data categories and example VQA questions.

Figure 2 : The Rho flow-matching action expert. State and noisy-action tokens pass through 12 wide transformer blocks with grouped-query self-attention, cross-attention to the VLM context, and feed-forward updates. The flow timestep conditions the sublayers through shared adaLN-Zero maps with block-specific offsets.

Component	Specification
Action expert	542M parameters
Transformer	12 blocks; width 2,048; MLP width 4,096
Attention	16 query heads; 4 key/value heads; head dimension 128
Block structure	Action self-attention; cross-attention to VLM context; feed-forward network
Flow conditioning	Shared adaLN-single with block-specific offsets
Flow target	Linear noise–action interpolation; velocity-prediction MSE

Table 3: Configuration of the Rho flow-matching action expert.

Panel of Figure 3 — Figure 3 : Rho-Tomyum mixture. Its 90% is composed of demonstration trajectories from physical and simulated robots as well as UMI-style demonstrations collected with standard YAM grippers, and the remaining 10% are vision-language data. All the robot/UMI-style data is bimanual, with the exception of a small subset of Open X-Embodiment Magic Soup [ 4 ] we call OXE Amuse Bouche . The robot data includes trajectories from policies controlling a UR AI Trainer simulated in NVIDIA Isaac Sim, trained using RL and the OmniReset framework [ 63 ] .

Panel of Figure 4 — Figure 4 : Midtraining mixtures of the embodiment-specific Rho variants. The count of distinct language instructions includes the number of both task and subtask descriptions.

Variant	Width	Heads	Head dim.	Blocks	Success
1,024 / 16 heads / 16 blocks	1,024	16	64	16	0.617
1,024 / 8 heads / 16 blocks	1,024	8	128	16	0.640
1,024 / 8 heads / 32 blocks	1,024	8	128	32	0.642
2,048 / 8 heads / 16 blocks	2,048	8	256	16	0.623
2,048 / 16 heads / 16 blocks	2,048	16	128	16	0.673

Table 4: Action-expert shape sweep on RoboEval after 40k adaptation steps. Success rates are averaged over eight tasks and three evaluation passes.

Variant	Q/KV heads	Blocks	AdaLN	AE params.	Success
Wide baseline	16/16	16	per-block	1.649B	0.673
Shared AdaLN	16/16	16	shared	882M	0.673
Grouped-query	16/4	16	shared	693M	0.672
Rho	16/4	12	shared	542M	0.671

Table 5: Action-expert compression on RoboEval while preserving width 2,048 and head dimension 128. Success rates are averaged over eight tasks after 40k adaptation steps. Each variant uses three evaluation passes.

Panel of Figure 5 — Figure 5 : Pretraining on 15% of the dataset mixture improves downstream adaptation. Left: LIBERO suite success after 20k task-adaptation steps. Right: RoboEval success throughout adaptation. The robot+VL condition combines robot trajectories with physical-grounding examples, the robot-data-only condition removes the physical-grounding batches, and the no-pretraining condition begins task adaptation from the grounded backbone with a randomly initialized action expert. Values are means and standard deviations over three evaluations, except the robot+VL RoboEval result at 40k adaptation steps, which pools six passes from two independent evaluations.

Pretraining learning rate	20k adaptation	30k adaptation	40k adaptation
$5\times 10^{-5}$	$0.601\pm 0.028$	$0.645\pm 0.005$	$0.641\pm 0.006$
$1\times 10^{-4}$	$\mathbf{0.623}\pm 0.031$	$\mathbf{0.675}\pm 0.035$	$\mathbf{0.673}\pm 0.015$
$2\times 10^{-4}$	$0.586\pm 0.010$	$0.629\pm 0.031$	$0.633\pm 0.008$
$5\times 10^{-4}$	$0.548\pm 0.006$	$0.637\pm 0.021$	$0.622\pm 0.022$

Table 6: Effect of the action-expert learning rate during pretraining on downstream RoboEval adaptation. All rows are pretrained on the same 15% subset of the dataset mixture and include physical-grounding data. The results are mean success and standard deviation over three evaluation passes.

Panel of Figure 6 — Figure 6 : Embodiment midtraining on RoboTwin transfers to held-out tasks across downstream-data budgets. Each checkpoint is adapted for 40k steps and evaluated on the matched five-task protocol with 50 episodes per task. Bars are means over completed seeds (four for Easy, three for Hard) and error bars are sample standard deviations. The benefit of midtraining is largest in the low-data regime and shrinks as the adaptation set grows.

Panel of Figure 7 — Figure 7 : FR3 Duo horizontal midtraining improves subsequent vertical adaptation to bimanual plug insertion. Every condition uses the same 50k-step task-adaptation budget; success reported over 30 trials.

Panel of Figure 8 — Figure 8 : Success rates of finetuned Rho, $\pi_{0.5}$ , GR00T-N1.7, and MolmoAct2 on RoboEval tasks. Rho outperforms all other models on Pick Book , Stack Book Shelf , and Rotate Valve . On each of the remaining 5 tasks, Rho matches the performance of the best baseline on that task. Importantly, each task’s best-performing baseline is generally different, so the next-strongest model, $\pi_{0.5}$ , matches Rho’s performance on only 4 out of 8 tasks.

Model	Succ. $\uparrow$	TP $\uparrow$	BGVD $\downarrow$	CPL $\downarrow$	ECC $\downarrow$	JPL $\downarrow$	OPL $\downarrow$	SCC $\downarrow$	SC $\downarrow$
MolmoAct2	0.61	0.79	0.072	1.48	1.03	7.97	4.84	0.017	0.465
GR00T N1.7	0.61	0.80	0.078	1.66	1.25	9.64	5.43	0.095	0.540
$\pi_{0.5}$	0.67	0.82	0.074	1.45	0.99	7.98	4.82	0.038	0.482
Rho	0.73	0.85	0.073	1.41	1.02	7.70	4.69	0.035	0.477

Table 7: Aggregate RoboEval outcome and behavioral metrics after 40k task-adaptation steps with global batch size 128. Rho outperforms the baselines on 5 out of the 9 metrics, with the next-strongest model across these metrics, MolmoAct2, winning on 3. Every row is the mean of 3 800-episode evaluations across the 8 task variations. Joint Path length (JPL) measures the sum of joint path lengths traced by the two arms. Arrows indicate the preferred direction of each metric, e.g., “

\downarrow

” means that lower is better.

Model	Spatial	Object	Goal	Long	Average
OpenVLA [ 23 ]	84.7	88.4	79.2	53.7	76.5
$\pi_{0}$ [ 4 ]	96.8	98.8	95.8	85.2	94.2
MolmoAct-7B-D [ 29 ]	87.0	95.4	87.6	77.2	86.6
GR00T N1.7 [ 42 ]	95.0	100.0	98.0	93.0	96.5
$\pi_{0.5}$ [ 48 ]	98.8	98.2	98.0	92.4	96.9
MolmoAct2 [ 14 ]	97.8	100.0	97.8	93.2	97.2

Table 8: Rho’s mean performance and published success rates of several baselines on the 4 task suites of LIBERO [ 34 ] . Rho outperforms the baselines in overall success rate aggregated across all 4 task suites and in success rate on the most challenging suite, LIBERO-long . Baseline values are taken from the corresponding model report or official release as cited in column 1; evaluation and adaptation protocols may differ across the reports. For Rho, the mean success rate is over 3 seeds. All values are percentages. Bold denotes the best result in each column.

Panel of Figure 9 — Figure 9 : The 6 BusyBox task categories used in the YAM Box experiments. Bimanual tasks require two-arm coordination; dual-arm tasks require choosing the correct arm for task completion.

Panel of Figure 10 — Figure 10 : YAM Box BusyBox results. After task-specific finetuning, Rho-YAM-Box achieves the same 90% mean success rate as $\pi_{0.5}$ and a higher mean success rate than GR00T-N1.7 and MolmoAct2 across 60 matched evaluation rollouts.

Panel of Figure 11 — Figure 11 : The UR AI Trainer tasks and their stages. Each row shows task progression from left to right.

Panel of Figure 12 — Figure 12 : UR AI Trainer results. Rho-UR-AI-Trainer achieves higher success than both baselines on both tasks. On Small-electronics cleanup , $\pi_{0.5}$ achieves similar mean task progress (80.8% versus 82.5%) despite lower success (40.0% versus 60.0%). $\pi_{0.5}$ ’s failures tend to occur during final placement, whereas Rho’s failures more often occur during picking.

Panel of Figure 13 — Figure 13 : The FR3 Duo tasks and their stages. The Test-tube and Tumbler tasks are evaluated in both directions: their frames read from left to right for the task on the upper arrow and from right to left for the task on the lower arrow.

Panel of Figure 14 — Figure 14 : FR3 Duo task-adaptation results. Each model is adapted independently on each task dataset and evaluated over 30 distinct, matched initial configurations per variant. We report binary success rate and mean task progress; Overall is the unweighted mean across the five task variants.

Task	Offline	+ Online	$\Delta$
Assembly	45.0	78.0	+33.0
Stick pull	58.0	78.0	+20.0
Pick out of hole	22.0	37.0	+15.0
Hand insert	73.0	86.0	+13.0
Hammer	92.0	100.0	+8.0
Basketball	55.0	61.0	+6.0

Table 9: Online latent-space adaptation on Meta-World. The offline policy is Rho trained with a limited budget on all 50 Meta-World tasks; the online column adapts it to each task in latent space from 50 episodes supervised by a privileged scripted expert. Success is measured over 100 rollouts per task; all values are percentages.

	Offline		+ Online
Task	SR	TP	SR	TP
Test-tube assembly	30.0	56.7	70.0	81.7
Plug insertion	66.7	83.3	93.3	96.7

Table 10: Online latent-space adaptation on FR3 Duo. Offline policies are adapted from the Rho-FR3-Duo checkpoint on 150 demonstrations per task; online adaptation uses 15 additional human-supervised episodes. Each task is evaluated over 30 trials, three on each of 10 hard initial configurations. SR is success rate and TP is task progress; both are percentages.

Appendix figures & tables12 assets

Supplementary material from the paper’s appendix.

Appendix

Component	Configuration
Vision-language backbone	Phi-Phy; 4.68B parameters
Language decoder	32 causal-transformer blocks; width 3,072; MLP width 8,192; 32 query and 32 key/value heads
Text interface	Vocabulary size 100,352; configured context length 16,384
Vision encoder	SigLIP 2 SO400M NaFlex; 428M parameters; $16\!\times\!16$ patches; penultimate-layer visual features
Visual-token budget	256–3,600 tokens per image; robot images are resized with padding to $256\!\times\!256$ , yielding 256 visual patches per view
Cross-modal projector	Two-layer MLP, $1{,}152\!\rightarrow\!3{,}072\!\rightarrow\!3{,}072$ , with GELU

Appendix

Table 11: Phi-Phy backbone configuration used by Rho.

Component	Configuration
Action stream	One projected state token followed by one noisy-action token for every position in the action chunk
Transformer	12 blocks; width 2,048; feed-forward width 4,096; GELU activation; dropout 0.1 during pretraining
Block structure	Action-stream self-attention, cross-attention to the fixed VLM context, then a feed-forward update, each with a gated residual
Attention	16 attention heads, 4 key/value heads, head dimension 128; grouped-query attention is used in both attention modules
Normalization	DiT-style adaLN-Zero; shared adaLN-single base maps with zero-initialized block- and site-specific offsets
Position encoding	Fixed sinusoidal encoding; configured maximum sequence length 6,144

Appendix

Table 12: Configuration of the Rho flow-matching action expert.

Signal	Source representation	Mapping	Use in action expert
Visual-language context	Block-14 Phi-Phy hidden sequence, width 3,072	Learned linear $3{,}072\!\rightarrow\!2{,}048$	Shared as the key/value sequence for cross-attention in every expert block
Robot state	One state vector padded to width 32	Learned linear $32\!\rightarrow\!2{,}048$	Prepended to the action-query sequence as one token
Noisy action	$H\!\times\!32$ flow state	Linear action projection followed by a $4{,}096\!\rightarrow\!2{,}048\!\rightarrow\!2{,}048$ action–time MLP	Supplies the remaining $H$ query tokens
Flow time	Scalar $t$ with a 2,048-dimensional sinusoidal embedding	$2{,}048\!\rightarrow\!2{,}048\!\rightarrow\!2{,}048$ SiLU MLP	Conditions the shared adaLN maps in every block
Padding masks	Valid image/text tokens, state/action dimensions, and chunk positions	No learned mapping	Masks attention or loss terms associated with padding

Appendix

Table 13: The interface between Phi-Phy and the continuous action expert. The backbone context is computed once per observation and reused by all flow integration steps.

Component	Parameters	Pretraining status
Phi-Phy vision–language backbone	4.68B	Trainable except token embeddings
SigLIP 2 vision encoder	428M	Trainable; included above
Token-embedding table	308M	Frozen; included above
Action expert	542M	Trainable from random initialization
Complete Rho model	5.22B	Approximately 4.91B trainable

Appendix

Table 14: Parameter accounting for Rho. Counts are rounded to the precision used throughout the paper.

Inference stage	Operation
Observation encoding	Encode all camera views and the instruction with Phi-Phy, extract decoder block 14, and project the valid context tokens to width 2,048. This backbone pass occurs once per policy query.
Latent initialization	Draw an independent Gaussian tensor $x_{1}\in\mathbb{R}^{H\times 32}$ ; an internal learned latent policy can replace this draw during online adaptation ( Appendix F ).
Flow integration	Apply 10 uniform explicit-Euler steps from $t=1$ to $t=0$ . Each step re-embeds the current action latent and queries the 12-block action expert while reusing the cached VLM context.
Action decoding	Project the final action tokens to 32 dimensions, remove padded dimensions, and invert the embodiment-specific normalization. No autoregressive text decoding is used to produce robot actions.
Receding-horizon execution	Execute a configurable prefix and then replan. Multi-embodiment pretraining uses $H=50$ and a 25-action execution horizon; RoboEval uses 32/16 and LIBERO uses 16/8.
Numerics	Bfloat16 model execution with FlashAttention 2; the Euler state update is accumulated in float32 before conversion back to bfloat16.

Appendix

Table 15: Rho inference path. Prediction and execution horizons are adapted to the control frequency and temporal extent of each embodiment.

Component	Configuration
Initialization	Phi-Phy backbone; randomly initialized state/action projections and action expert
Trainable parameters	Vision encoder, cross-modal projector, language decoder, and action modules; token embeddings frozen
Robot inputs	One observation; up to three RGB views; padded $256\!\times\!256$ images; task instruction and proprioceptive state
Action targets	Chunk-relative EEF translation and 6D-rotation deltas; absolute grippers; approximately 1 s at the native rate; at most 50 steps; per-position 1st/99th-percentile normalization
Training objectives	Flow matching on robot batches; autoregressive cross-entropy on VQA and pointing batches, weighted by $0.02$
Batch sampling	Robot/VL update probability $0.9/0.1$ ; global batch 3,072/384

Appendix

Table 16: Multi-embodiment pretraining recipe for Rho. Batch sizes count examples per modality-homogeneous optimization update.

Component	Configuration
Initialization	First-epoch multi-embodiment Rho checkpoint
Data mixture	FR3 Duo multi-task robot data and retained vision–language data, sampled at a 9:1 robot/VL ratio
Trainable parameters	Vision encoder, cross-modal projector, language decoder, and action modules; token embeddings frozen
Action representation	One-second chunk of chunk-relative Cartesian end-effector translation and 6D-rotation deltas with absolute gripper commands, padded to 50 positions
Batch and updates	128 examples per GPU on 8 B200 GPUs (global batch 1,024); 165,000 optimizer updates
Optimizer	AdamW; $\beta_{1}=0.9$ , $\beta_{2}=0.95$ ; $\epsilon=10^{-6}$ ; weight decay $0.1$

Appendix

Table 17: FR3 Duo embodiment-midtraining recipe.

Component	Configuration
Initialization	FR3 Duo embodiment-midtrained Rho checkpoint
Data	Target-task demonstrations only; three RGB views, language, and proprioception
Trainable parameters	Vision encoder, cross-modal projector, language decoder, and action modules; token embeddings frozen
Action representation	One-second chunk of chunk-relative Cartesian end-effector translation and 6D-rotation deltas with absolute gripper commands, padded to a 50-position model chunk
Normalization	Per-task action-chunk mean and standard deviation
Batch and updates	16 examples per GPU on 4 B200 GPUs (global batch 64); 50,000 optimizer updates

Appendix

Table 18: Representative FR3 Duo task-finetuning recipe.

Benchmark	Tasks	Adaptation	Trials / task / pass	Evaluation passes
LIBERO	40 (four suites)	40k updates	50	3
RoboEval	8	40k updates	100	3
RoboTwin	5 (each under Easy & Hard settings)	40k updates	50	3 (Hard)/4 (Easy)

Appendix

Table 19: LIBERO, RoboEval, and RoboTwin task-adaptation and evaluation budgets for the reported Rho results. Each pass is a complete evaluation over the corresponding task set.

Suite	Tasks
Spatial	Move the black bowl to the plate from: between the plate and ramekin; next to the ramekin; the table center; atop the cookie box; the cabinet’s top drawer; atop the ramekin; next to the cookie box; the stove; next to the plate; and atop the wooden cabinet.
Object	Place in the basket: alphabet soup; cream cheese; salad dressing; BBQ sauce; ketchup; tomato sauce; butter; milk; chocolate pudding; and orange juice.
Goal	Open the cabinet’s middle drawer; put the bowl on the stove; put the wine bottle atop the cabinet; open the top drawer and put the bowl inside; put the bowl atop the cabinet; push the plate in front of the stove; put the cream cheese in the bowl; turn on the stove; put the bowl on the plate; put the wine bottle on the rack.
Long	Put alphabet soup and tomato sauce in the basket; put cream cheese and butter in the basket; turn on the stove and put the moka pot on it; put the black bowl in the bottom drawer and close it; put the white mug on the left plate and the yellow-and-white mug on the right plate; put the book in the caddy’s rear compartment; put the white mug on the plate and the chocolate pudding to its right; put alphabet soup and cream cheese in the basket; put both moka pots on the stove; put the yellow-and-white mug in the microwave and close it.

Appendix

Table 20 : LIBERO task inventory used in our simulation evaluation. Spatial and Object hold the goal fixed while varying the relevant spatial relation or object; Long contains the benchmark’s ten multi-stage tasks.

Hyperparameter	Meta-World	FR3 Duo
Correction source	Scripted expert	Human (SpaceMouse)
Online episodes per task	50	15
Action/latent horizon	16	16
Latent dimension per step	32	32
Latent bound	3.0	3.0
Inversion method	Per-step fixed point	Per-step fixed point

Appendix

Table 21: Online latent-space adaptation hyperparameters. Rho’s vision-language backbone and flow-matching action expert are frozen in both settings; only its lightweight internal latent policy is optimized.

Task	Model	Succ. $\uparrow$	TP $\uparrow$	BGVD $\downarrow$	CPL $\downarrow$	ECC $\downarrow$	JPL $\downarrow$	OPL $\downarrow$	SCC $\downarrow$	SC $\downarrow$
Cube handover	GR00T N1.7	0.64	0.93	0.043	1.35	0.16	6.81	4.14	0.373	0.477
	$\pi_{0.5}$	0.79	0.92	0.038	1.18	0.10	5.44	3.37	0.210	0.237
	MolmoAct2	0.86	0.95	0.036	1.11	0.04	5.12	3.30	0.107	0.193
	Rho	0.80	0.93	0.037	1.18	0.09	5.15	3.14	0.207	0.217
Lift pot	GR00T N1.7	0.67	0.81	0.058	2.57	0.13	16.83	9.70	0.120	0.603
	$\pi_{0.5}$	0.72	0.84	0.029	1.48	0.00	9.39	6.80	0.017	0.417

Appendix

Table 22 : Task-level RoboEval outcome and behavioral metrics after 40k task-adaptation steps. Each cell is pooled over three 100-episode evaluations for the corresponding task. Path-length measures sum the two arms. Arrows indicate the preferred direction.

Explore similar work

PriorVLA: Prior-Preserving Adaptation for Vision-Language-Action Models
May 11, 2026Xinyu Guo, Bin Xie, Wei Chai +4Diffusion-Based Vision-Language-Actions Vision-Language Model Adaptation
Institute of Automation, Chinese Academy of Sciences · Dexmal · Nanjing University of Aeronautics and Astronautics +1
Agentic-VLA: Efficient Online Adaptation for Vision-Language-Action Models
May 21, 2026Ruofan Jin, Zaixi ZhangDiffusion-Based Vision-Language-Actions Vision-Language Model Adaptation
Scinetics.
Robotic Policy Adaptation via Weight-Space Meta-Learning
Jun 5, 2026Christian Bianchi, Siamak Yousefi, Alessio Sampieri +4Robot Policies Robotic Manipulation
ItalAI · ItalAI and Sapeinza University of Rome

May 11, 2026, cs.ROJ/K move · Enter open · S save

PriorVLA: Prior-Preserving Adaptation for Vision-Language-Action Models

Xinyu Guo, Bin Xie, Wei Chai, Xianchi Deng+3

Institute of Automation, Chinese Academy of Sciences · Dexmal · Nanjing University of Aeronautics and Astronautics+1

Diffusion-Based Vision-Language-Actions Vision-Language Model Adaptation Robotic Manipulation Pretraining Visual Priors

Open paper AskPDF arXiv

Large-scale pretraining has made Vision-Language-Action (VLA) models promising foundations for generalist robot manipulation, yet adapting them to downstream tasks remains necessary. However, the common practice of full fine-tuning treats pretraining as initialization and can shift broad priors toward narrow training-distribution patterns. We propose PriorVLA, a novel framework that preserves pretrained priors and learns to leverage them for effective adaptation. PriorVLA keeps a frozen Prior Expert as a read-only prior source and trains an Adaptation Expert for downstream specialization. Expert Queries capture scene priors from the pretrained VLM and motor priors from the Prior Expert, integrating both into the Adaptation Expert to guide adaptation. Together, PriorVLA updates only 25% of the parameters updated by full fine-tuning. Across RoboTwin 2.0, LIBERO, and real-world tasks, PriorVLA achieves stronger overall performance than full fine-tuning and state-of-the-art VLA baselines, with the largest gains under out-of-distribution (OOD) and few-shot settings. PriorVLA improves over pi0.5 by 11 points on RoboTwin 2.0-Hard and achieves 99.1% average success on LIBERO. Across eight real-world tasks and two embodiments, PriorVLA reaches 81% in-distribution (ID) and 57% OOD success with standard data. With only 10 demonstrations per task, PriorVLA reaches 48% ID and 32% OOD success, surpassing pi0.5 by 24 and 22 points, respectively.

Abstract

Explore similar work

PriorVLA: Prior-Preserving Adaptation for Vision-Language-Action Models

Agentic-VLA: Efficient Online Adaptation for Vision-Language-Action Models

Robotic Policy Adaptation via Weight-Space Meta-Learning