Pith. sign in

REVIEW 2 cited by

SuperSAM: Crafting a SAM Supernetwork via Structured Pruning and Unstructured Parameter Prioritization

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2501.08504 v1 pith:LPPWWKBH submitted 2025-01-15 cs.CV cs.LG

SuperSAM: Crafting a SAM Supernetwork via Structured Pruning and Unstructured Parameter Prioritization

classification cs.CV cs.LG
keywords searchspacedesignefficientone-shotparameterprioritizationpruning
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Neural Architecture Search (NAS) is a powerful approach of automating the design of efficient neural architectures. In contrast to traditional NAS methods, recently proposed one-shot NAS methods prove to be more efficient in performing NAS. One-shot NAS works by generating a singular weight-sharing supernetwork that acts as a search space (container) of subnetworks. Despite its achievements, designing the one-shot search space remains a major challenge. In this work we propose a search space design strategy for Vision Transformer (ViT)-based architectures. In particular, we convert the Segment Anything Model (SAM) into a weight-sharing supernetwork called SuperSAM. Our approach involves automating the search space design via layer-wise structured pruning and parameter prioritization. While the structured pruning applies probabilistic removal of certain transformer layers, parameter prioritization performs weight reordering and slicing of MLP-blocks in the remaining layers. We train supernetworks on several datasets using the sandwich rule. For deployment, we enhance subnetwork discovery by utilizing a program autotuner to identify efficient subnetworks within the search space. The resulting subnetworks are 30-70% smaller in size compared to the original pre-trained SAM ViT-B, yet outperform the pretrained model. Our work introduces a new and effective method for ViT NAS search-space design.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Lean-SAM2: Target-Anchored Memory and Encoder Acceleration for SAM2

    cs.CV 2026-07 conditional novelty 5.0

    Lean-SAM2 combines target-anchored memory pruning, condensed insurance memory, and risk-aware window routing to accelerate SAM2.1 inference ~1.4× with better accuracy than Efficient-SAM2.

  2. On Efficient Variants of Segment Anything Model: A Survey

    cs.CV 2024-10 unverdicted novelty 5.0

    A survey that reviews efficient variants of the Segment Anything Model, categorizes acceleration strategies, and provides a unified hardware evaluation on benchmarks.