Paper Detail

SLICEChat: Progressive In-Encoder Token Pruning for Whole-Slide Pathology Language Models

Ali Kerem Bozkurt, Baris Cem Bakay, Ibrahim Kulac, Cigdem Gunduz-Demir, Erkut Erdem, Aykut Erdem

arxiv Score 22.3

Published 2026-09-21 · First seen 2026-09-22

General AI

Abstract

Whole-slide pathology images (WSIs) contain gigapixel-scale visual content, creating a major scalability challenge for slide-level multimodal large language models (MLLMs). Existing approaches process thousands of patch tokens and typically apply compression only after slide encoding, leaving multimodal attention computationally expensive. We introduce SLICEChat, a slide-level MLLM that integrates progressive token pruning within a hybrid Mamba--Transformer slide encoder. Mamba layers enable efficient long-range propagation, while Transformer layers preserve global interactions as the sequence is progressively shortened. Between stages, language-supervised, region-aware pruning removes spatially coherent low-utility regions under a controlled keep-rate schedule, producing compact slide representations before multimodal fusion. On SlideBench VQA, SLICEChat achieves 79.84% accuracy on TCGA and 59.09% on BCNB cohorts, outperforming prior slide-level pathology MLLMs, and achieves the highest overall WSI-Bench metrics. It also provides competitive memory usage and the inference latency among the evaluated models. These results demonstrate accurate and computationally efficient multimodal reasoning over gigapixel WSIs.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
now
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@article{bozkurt2026slicechat,
  title = {SLICEChat: Progressive In-Encoder Token Pruning for Whole-Slide Pathology Language Models},
  author = {Ali Kerem Bozkurt and Baris Cem Bakay and Ibrahim Kulac and Cigdem Gunduz-Demir and Erkut Erdem and Aykut Erdem},
  year = {2026},
  abstract = {Whole-slide pathology images (WSIs) contain gigapixel-scale visual content, creating a major scalability challenge for slide-level multimodal large language models (MLLMs). Existing approaches process thousands of patch tokens and typically apply compression only after slide encoding, leaving multimodal attention computationally expensive. We introduce SLICEChat, a slide-level MLLM that integrates progressive token pruning within a hybrid Mamba--Transformer slide encoder. Mamba layers enable eff},
  url = {https://arxiv.org/abs/2609.24894},
  keywords = {cs.CV, cs.CL},
  eprint = {2609.24894},
  archiveprefix = {arXiv},
}

Metadata

{}