AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Mimicking Radiologists: A Coarse-to-Fine Framework with Structural Sparse Tokens for Dual-LLM Computed Tomography Report Generation

Hong Liu, Dong Wei, Yefeng Zheng, Xian Wu, Liansheng Wang

ECCV 2026 · Architectures and efficiency › Attention and transformers · approach: Block-sparse attention kernels · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add mask-guided sparse negative-entropy loss (ablation)baseline CTRG · CT-RATE (test)CE F1: 0.356 -> 0.377Improved
add mask-guided sparse negative-entropy loss (ablation)baseline CTRG · CT-RATE (test)BLEU-1: 0.379 -> 0.375Hurt
add abnormality-prefixed SFT (ablation)baseline + L_sparse · CT-RATE (test)CE F1: 0.377 -> 0.394Improved
add abnormality-prefixed SFT (ablation)baseline + L_sparse · CT-RATE (test)BLEU-4: 0.192 -> 0.282Improved
add AP-LTF (K=200 filter) without GRPO (ablation)baseline + L_sparse + AP-SFT · CT-RATE (test)CE F1: 0.394 -> 0.391Hurt
add AP-LTF (K=200 filter) without GRPO (ablation)baseline + L_sparse + AP-SFT · CT-RATE (test)CE recall: 0.397 -> 0.418Improved
add GRPO collaborative refinement (ablation)baseline + L_sparse + AP-SFT + AP-LTF · CT-RATE (test)CE F1: 0.391 -> 0.415Improved
add GRPO collaborative refinement (ablation)baseline + L_sparse + AP-SFT + AP-LTF · CT-RATE (test)BLEU-4: 0.289 -> 0.305Improved
use two specialized LLMs instead of single multitask LLM (ablation)Ours · CT-RATE (test)CE F1: 0.381 -> 0.415Improved
use two specialized LLMs instead of single multitask LLM (ablation)Ours · CT-RATE (test)BLEU-4: 0.302 -> 0.305Hurt
use composite F1+BLEU-4 reward instead of BLEU-4 alone (ablation)GRPO refinement · CT-RATE (test)CE F1: 0.401 -> 0.415Improved
use composite F1+BLEU-4 reward instead of F1 alone (ablation)GRPO refinement · CT-RATE (test)CE F1: 0.408 -> 0.415Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy