# The SAM2-to-SAM3 Gap in the Segment Anything Model Family: Why Prompt-Based Expertise Fails in Concept-Driven Image Segmentation

Source: [The SAM2-to-SAM3 Gap in the Segment Anything Model Family: Why Prompt-Based Expertise Fails in Concept-Driven Image Segmentation](<https://www.sota2.com/research/paper/6937ad196cfd488761a4e88b>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-12-12T06:00:01.695Z

Authors: Ranjan Sapkota, Konstantinos I. Roumeliotis, Manoj Karkee

Published: 2025-12-04T16:27:18.000Z

## Abstract

This paper investigates the fundamental discontinuity between the latest two Segment Anything Models: SAM2 and SAM3. We explain why the expertise in prompt-based segmentation of SAM2 does not transfer to the multimodal concept-driven paradigm of SAM3. SAM2 operates through spatial prompts points, boxes, and masks yielding purely geometric and temporal segmentation. In contrast, SAM3 introduces a unified vision-language architecture capable of open-vocabulary reasoning, semantic grounding, contrastive alignment, and exemplar-based concept understanding. We structure this analysis through five core components: (1) a Conceptual Break Between Prompt-Based and Concept-Based Segmentation, contrasting spatial prompt semantics of SAM2 with multimodal fusion and text-conditioned mask generation of SAM3; (2) Architectural Divergence, detailing pure vision-temporal design of SAM2 versus integration of vision-language encoders, geometry and exemplar encoders, fusion modules, DETR-style decoders, object queries, and ambiguity-handling via Mixture-of-Experts in SAM3; (3) Dataset and Annotation Differences, contrasting SA-V video masks with multimodal concept-annotated corpora of SAM3; (4) Training and Hyperparameter Distinctions, showing why SAM2 optimization knowledge does not apply to SAM3; and (5) Evaluation, Metrics, and Failure Modes, outlining the transition from geometric IoU metrics to semantic, open-vocabulary evaluation. Together, these analyses establish SAM3 as a new class of segmentation foundation model and chart future directions for the emerging concept-driven segmentation era.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2512.06032>)

[PDF](<https://arxiv.org/pdf/2512.06032.pdf>)

[Implementation](<https://github.com/Applied-AI-Research-Lab/The-SAM2-to-SAM3-Gap-in-the-Segment-Anything-Model-Family>)
