aiqbench analysis
Slot2Text introduces an efficient visual tokenization method for surgical multimodal large language models by replacing dense image tokens with compact region-based "slots." This approach significantly reduces inference costs while offering two modes: one optimized for speed and another that provides explicit spatial grounding for reasoning tasks. The work advances the practical deployment of AI in medical imaging by balancing computational efficiency with traceable visual evidence.
Slot2Text: Object-Centric Visual Tokenization for Efficient and Spatially Traceable Surgical MLLMs
arXiv:2608.01473v1 Announce Type: cross Abstract: Multimodal large language models (MLLM) for surgical scene understanding typically inject hundreds of dense visual tokens into a language model, leading to costly inference and limited spatial traceability for generated answers. We present Slot2Text, a dual-mode surgical MLLM that replaces dense representations of visual input with a compact set of regions encoded as slot latents. Instead of relying on contrastive alignment of the visual encoder with language, Slot2Text groups self-supervised vision features into a few regions--slots that are consumed by the language model as area-labeled visual tokens. Slot2Text-Fast uses the slot prefix to answer surgical questions. Slot2Text-Reason also identifies and locates areas relevant for reasoning, linking language outputs to corresponding slot tokens, masks or regions. Experiments on multiple visual question answering and visual grounding benchmarks show that Slot2Text-Fast is competitive with state-of-the-art baseline at a much lower cost, reducing the average total token consumption by a 91.8\% and the visual prefix from 1,295 to 47 tokens (a 96.4\% reduction). Slot2Text-Reason trades additional tokens and latency for explicit area identities, locations, and traceable spatial evidence. These results establish compact slot latents as an efficient default visual interface for surgical MLLMs, with grounded reasoning invoked when greater spatial traceability is required.
Abstract:Multimodal large language models (MLLM) for surgical scene understanding typically inject hundreds of dense visual tokens into a language model, leading to costly inference and limited spatial traceability for generated answers. We present Slot2Text, a dual-mode surgical MLLM that replaces dense representations of visual input with a compact set of regions encoded as slot latents. Instead of relying on contrastive alignment of the visual encoder with language, Slot2Text groups self-supervised vision features into a few regions--slots that are consumed by the language model as area-labeled visual tokens. Slot2Text-Fast uses the slot prefix to answer surgical questions. Slot2Text-Reason also identifies and locates areas relevant for reasoning, linking language outputs to corresponding slot tokens, masks or regions. Experiments on multiple visual question answering and visual grounding benchmarks show that Slot2Text-Fast is competitive with state-of-the-art baseline at a much lower cost, reducing the average total token consumption by a 91.8\% and the visual prefix from 1,295 to 47 tokens (a 96.4\% reduction). Slot2Text-Reason trades additional tokens and latency for explicit area identities, locations, and traceable spatial evidence. These results establish compact slot latents as an efficient default visual interface for surgical MLLMs, with grounded reasoning invoked when greater spatial traceability is required.
Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL); Machine Learning (cs.LG)
arXiv:2608.01473 [cs.CV]
(or arXiv:2608.01473v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2608.01473 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 303 words
Related coverage
Action with Visual Primitives
arXiv · Sep 22, 2026
Action with Visual Primitives
arXiv · Jun 16, 2026
Universal Pose Pretraining for Generalizable Vision-Language-Action Policies
arXiv · Sep 30, 2026
Vertical Advantage: Transforming Industries with Lakebase and Agentic AI
Databricks · Aug 27, 2026