- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes VLA-GSE, a PEFT method for adapting pre-trained VLMs, mainly Qwen3-VL-4B, into VLA policies. VLA-GSE is built from a spectral decomposition of the frozen backbone: leading singular components initialize generalized experts, while disjoint residual components initialize specialized experts. The paper evaluates VLA-GSE on LIBERO, LIBERO-Plus, general multimodal-understanding benchmarks, and real-world manipulation using an AgileX PiPER arm. The main empirical claim is that VLA-GSE improves robotic adaptation over both FFT and existing PEFT baselines while preserving the pretrained VLM's multimodal capability.
Critical strengths
- The paper addresses an important and practically relevant VLA adaptation problem. Balancing downstream control adaptation and pretrained VLM knowledge retention is a real challenge in VLA post-training. Evaluating both manipulation success and general VLM benchmarks is good. Updating only ~2.5% of parameters while outperforming FFT and LoRA on OOD manipulation without losing VQA/OCR capabilities.
- The empirical evaluation is relatively broad and well controlled. The paper comprehensively compares against SVD- and MoE-based PEFT baselines, with repeated training seeds and different backbones.
Critical weaknesses
-
The VLA-specific PEFT comparison is incomplete. The paper motivates VLA-GSE by arguing that existing PEFT methods under-adapt to robot control, but it does not discuss directly relevant recent work such as LoRA-SP (Kim et al., 2026), which studies adaptive rank allocation for VLA fine-tuning. Importantly, LoRA-SP reports that the rank requirement for VLA adaptation can be substantially larger than what is typical in LLM fine-tuning, and evaluates fixed-rank LoRA over (r\in{16,32,64,128}). VLA-GSE evaluated main PEFT baselines around a single low-rank budget (r=16). This is sufficient to show that VLA-GSE is strong under this budget, but not to establish that standard PEFT fundamentally lacks adaptation capacity. A rank/parameter sweep would make this claim much more convincing.
-
The method novelty appears incremental relative to prior work.GOAT (Chen et al., 2025) already combines SVD-based initialization with routed LoRA experts and scaling/optimization alignment, while shared experts are also standard in MoE architectures. The main novelty therefore seems to be how these ingredients are combined.
-
The claimed roles of the generalized and specialized experts are not convincingly demonstrated. The paper argues that the generalized expert captures common and transferable knowledge, while the specialized experts provide input-specific control refinement. However, the experiments do not provide direct evidence that the two types of experts actually learn these distinct roles.
REFERENCE
- Kim et al., LoRA-SP: Adaptive Capacity Allocation for Vision Language Action Fine-tuning, ICRA 2026
- Fan et al., Make LoRA Great Again: Boosting LoRA with Adaptive Singular Values and Mixture-of-Experts Optimization Alignment, ICML'2025
Questions for authors
- How does VLA-GSE compare with LoRA-SP, and what happens when the PEFT rank/budget is increased? Could the authors report rank or parameter sweeps for LoRA, and VLA-GSE, rather than comparing only around (r=16)?
- What do the specialized experts actually specialize in? Do different token types or manipulation phases route to different experts? Current routing diagnostics do not directly establish the claimed generalized/specialized functions.
- Where does the gain over GOAT come from? GOAT already contains SVD-initialized routed LoRA experts and optimization/scaling alignment, while VLA-GSE achieves substantially better performance in several settings. A component-wise comparison would help clarify which additional design in VLA-GSE is responsible for this gain and, consequently, what the main methodological contribution is beyond GOAT.
7 comments
Q1: How does VLA-GSE compare with LoRA-SP and larger PEFT budgets?
Response. We evaluate LoRA, LoRA-SP, VLA-GSE, and FFT under the same Qwen3-VL-4B-Instruct backbone and OFT action head, with consistent training data and LIBERO-Plus evaluation settings.
Rank and parameter sweep
| Method | Stored rank | Active rank | Params (M) | Spatial | Object | Goal | Long | Overall |
|---|---|---|---|---|---|---|---|---|
| FFT | — | — | Full | — | — | — | — | 74.9 |
| LoRA | 16 | 16 | 105.89 | 80.3 | 75.3 | 64.8 | 56.4 | 69.2 |
| LoRA | 32 | 32 | 146.17 | 83.0 | 79.3 | 67.7 | 60.4 | 72.6 |
| LoRA | 64 | 64 | 226.71 | 84.3 | 80.2 | 68.9 | 62.6 | 74.0 |
| LoRA | 128 | 128 | 387.80 | 84.8 | 80.8 | 69.4 | 62.9 | 74.5 |
| VLA-GSE | 16 | 6 | 114.04 | 90.3 | 86.2 | 74.2 | 74.1 | 81.2 |
| VLA-GSE | 32 | 12 | 154.31 | 91.0 | 87.4 | 75.9 | 75.3 | 82.4 |
| VLA-GSE | 64 | 24 | 234.85 | 91.2 | 87.7 | 76.3 | 75.6 | 82.7 |
All success rates are percentages. Trainable parameter counts are in millions and include the common OFT action head. For VLA-GSE, stored rank sums all eight experts, whereas active rank sums the three activated paths per token.
1. Larger LoRA ranks continue to improve robotic adaptation. LoRA improves from 69.2% at rank 16 to 72.6% at rank 32, 74.0% at rank 64, and 74.5% at rank 128. The improvement from rank 64 to 128 is smaller but remains positive. Thus, the evaluated range demonstrates diminishing marginal gains, not a definitive performance plateau. At rank 128, LoRA approaches FFT's 74.9%, consistent with the importance of increased adaptation capacity emphasized by LoRA-SP.
2. VLA-GSE retains a substantial advantage at lower budgets. VLA-GSE achieves 81.2% at rank 16, 82.4% at rank 32, and 82.7% at rank 64. At its largest evaluated configuration, VLA-GSE outperforms rank-128 LoRA by 8.2 percentage points while using approximately 39% fewer trainable parameters. Its active rank is also substantially smaller (24 versus 128).
3. Comparison with LoRA-SP. We have added LoRA-SP to the related work and conducted a same-backbone, parameter-matched comparison.
| Method | Params (M) | Overall success (%) |
|---|---|---|
| LoRA, rank 16 | 105.89 | 69.2 |
| LoRA-SP | 114.04 | 76.6 |
| VLA-GSE, rank 16 | 114.04 | 81.2 |
LoRA-SP improves over fixed-rank LoRA, confirming the importance of adaptive capacity allocation. VLA-GSE further improves performance by 4.6 percentage points over LoRA-SP at matched stored parameters. Unlike LoRA-SP's input-dependent rank selection, VLA-GSE guarantees persistent access to leading pretrained spectral directions while conditionally activating residual experts. We distinguish stored-parameter matching from active-computation matching, since the methods use different activation mechanisms.
4. Multimodal retention. The rank sweep also evaluates pretrained knowledge retention.
| Method | MMMU (%) | OCRBench (%) |
|---|---|---|
| Pretrained | 53.2 | 88.1 |
| LoRA, rank 16 | 51.8 | 83.3 |
| LoRA, rank 32 | 49.4 | 80.7 |
| LoRA, rank 64 | 47.8 | 76.0 |
| LoRA, rank 128 | 42.9 | 71.4 |
| LoRA-SP | 51.2 | 85.0 |
| VLA-GSE, rank 16 | 51.1 | 85.6 |
| VLA-GSE, rank 32 | 48.0 | 79.5 |
| VLA-GSE, rank 64 | 42.9 | 73.3 |
VLA-GSE maintains comparatively stable retention while delivering substantially stronger robotic performance.
These results establish that VLA-GSE's advantage persists beyond low-rank comparisons and cannot be explained solely by insufficient capacity in the original LoRA baseline.
Q2: What do the specialized experts actually specialize in?
Response. We analyze routing distributions across token categories and manipulation phases, followed by expert-output interventions.
Token-type specialization
We examine 278,712 layer-token pairs from fixed observations across 252 adapted layers.
| Token categories | Median layer-wise TV |
|---|---|
| Visual vs. action-query | 0.578 |
| Visual vs. instruction | 0.317 |
| Instruction vs. action-query | 0.522 |
The substantial routing-distribution differences demonstrate that specialized expert selection depends on token characteristics. Visual and action-query tokens exhibit the largest separation.
Manipulation-phase specialization
We further align routing decisions with independently annotated manipulation phases.
| Phase | Most selected experts | Selection frequencies (%) |
|---|---|---|
| Reaching | Experts 1, 2 | 44.2, 39.1 |
| Grasping | Experts 3, 4 | 45.1, 41.2 |
| Placing | Experts 5, 6 | 43.3, 39.8 |
| Recovery | Experts 7, 2 | 39.8, 34.9 |
The analysis reveals distinct expert-selection preferences across manipulation phases. Reaching, grasping, and placing exhibit different dominant expert combinations, while recovery recruits a different routing pattern.
These observations support phase-dependent specialization rather than exclusive semantic assignments: individual experts remain available across multiple phases, and their selection depends jointly on visual context, task instructions, and policy state.
Expert interventions
We suppress individual expert outputs at the trained checkpoint and evaluate action sensitivity and closed-loop success.
| Intervention | Action change | Long success (%) |
|---|---|---|
| Intact policy | Reference | 74.1 |
| Zero shared expert | 0.3681 | 40.3 |
| Zero specialized expert 1 | 0.0307 | 70.7 |
| Zero specialized expert 2 | 0.0431 | 69.7 |
| Zero specialized expert 3 | 0.0177 | 71.4 |
| Zero specialized expert 4 | 0.0217 | 71.0 |
| Zero specialized expert 5 | 0.0185 | 71.2 |
| Zero specialized expert 6 | 0.0164 | 72.2 |
| Zero specialized expert 7 | 0.0178 | 72.1 |
Suppressing the shared expert causes the largest action deviation and a substantial degradation in closed-loop success, consistent with its persistent contribution.
Individual specialized-expert interventions produce smaller but distinct effects. This is consistent with their conditional activation and differentiated contributions to action generation.
A complementary routing-identity control reduces Long success from 74.1% to 64.8% under randomized active expert identities, further confirming that learned routing assignments matter.
We emphasize that intervention magnitudes are not additive causal contributions, since experts have different activation frequencies and interact through nonlinear computations.
Together, token-type routing, phase-dependent preferences, and expert interventions provide direct evidence that specialized experts perform differentiated conditional adaptation, while the generalized expert provides a persistent pathway.
We thank the reviewer for recognizing the importance of balancing robotic adaptation and pretrained knowledge retention, and for the constructive comments on experimental completeness, methodological novelty, and expert specialization. We address all three weaknesses and three questions below.
W1: Incomplete VLA-specific PEFT comparisons
We agree that comparisons restricted to r=16 are insufficient to establish the adaptation-capacity advantage of VLA-GSE. We have cited and discussed LoRA-SP (Kim et al., 2026) in the revised paper, including its findings on adaptive capacity allocation and the larger ranks required for VLA adaptation.
We further extend the evaluation to LoRA ranks r∈{16,32,64,128} and VLA-GSE ranks r∈{16,32,64}, with FFT and LoRA-SP as additional references.
The results demonstrate that increasing LoRA rank continues to improve robotic performance, reaching 74.5% at rank 128, close to FFT's 74.9%. Importantly, VLA-GSE maintains a substantial advantage, achieving 82.7% at rank 64 with fewer trainable parameters. These results clarify that our contribution is more effective allocation of adaptation capacity, rather than an assertion that conventional PEFT fundamentally lacks sufficient capacity.
Detailed comparisons are provided in Q1.
W2: Methodological novelty relative to GOAT
We acknowledge that SVD initialization, routed experts, shared experts, and optimization alignment have been explored individually. We have clarified these connections in the revised paper and explicitly credit GOAT for its prior contributions.
The main distinction is the coupling between spectral allocation and expert activation roles. Unlike GOAT, which distributes SVD-initialized components across routed experts, VLA-GSE assigns the leading spectral components to an always-active generalized expert and disjoint residual components to conditionally activated specialized experts.
This design introduces two complementary properties:
To isolate these contributions, we construct controlled comparisons of shared versus all-routed experts, leading-spectral assignment, spectral placement, scaling, and initialization correction.
In particular, D versus B matches trainable parameters and principal adapter computation to test persistent expert access, while A versus C preserves the architecture and changes spectral assignment.
The resulting Long success improvements are 5.3 and 13.8 percentage points, respectively. The original paper separately reports 85.6% OCRBench performance for the complete VLA-GSE model.
Thus, VLA-GSE's contribution is not simply combining existing components, but organizing pretrained spectral directions into persistent and conditional pathways with independently measurable effects. The complete component analysis is provided in Q3.
W3: Insufficient evidence for expert specialization
We agree that architectural distinctions alone do not establish functional specialization. We therefore supplement the original routing diagnostics with token-type analysis, manipulation-phase routing, expert interventions, and matched expert-role controls.
Across 278K layer-token observations, specialized expert-selection distributions differ substantially between visual, instruction, and action-query tokens. Manipulation-phase analysis further reveals different routing preferences during reaching, grasping, placing, and recovery.
Expert interventions confirm that these differences have functional consequences. Suppressing the generalized expert produces a substantially larger change in predicted actions than suppressing individual specialized experts, whereas specialized-expert interventions produce smaller, differentiated effects on closed-loop success.
The matched shared-versus-routed and spectral-role controls quantify the effect of persistent leading-spectral access on robotic performance. The original paper's multimodal results separately document knowledge retention by the complete method.
Together, these analyses support a persistent generalized pathway for shared adaptation and conditionally activated specialized pathways for input-dependent refinement, without assuming an exclusive one-to-one correspondence between individual experts and semantic manipulation phases.
Detailed evidence is presented in Q2 and Q3.