9 papers
TwinPrune: Density-Aware Two-Phase Token Pruning for Vision-Language Models
Kai Liu, Anqi Li, Junxian Li +4
We propose a unified spatial-density signal to score token importance for both inference phases: bipartite merge for prefill and KV eviction for decode, yielding measurable gains i…
OpenTumorBoard: A Real-World Benchmark of Multidisciplinary Tumor Board Discussion Trajectories
Anqi Li, Zhixuan Ge, Yixuan Duan +9
Multidisciplinary tumor boards integrate multimodal clinical observations and longitudinal patient histories through specialist discussions, yet benchmarks rarely capture these rea…
OrganLens: Organ-Specific Representation Learning for CT Foundation Models
Zhixuan Ge, Anqi Li, Sadeer Al-Kindi +2
A CT examination captures multiple organs, but many biomedical questions concern abnormalities, prognosis, or longitudinal change in a specific organ. These questions require a sep…
UniNote: A Unified Embedding Model for Multimodal Representation and Ranking
Jinghan Zhao, Wenwei Jin, Anqi Li +5
Item-to-Item (I2I) retrieval is a fundamental part of modern content platforms, supporting critical industrial workflows from recommendation engines to content auditing. While mult…
DRScaffold: Boosting Dense-Scene Reasoning in Lightweight Vision Language Models
Xinrui Shi, Kai Liu, Ziqing Zhang +3
Lightweight vision-language models perform competitively on standard benchmarks yet fail systematically in dense-scene reasoning, where multiple objects, attributes, and relations…
MedMIX: Modality-Internal Expert Fusion for Multimodal Medical Diagnosis
Seungik Cho, Anqi Li, Wei Qiu
Multimodal clinical prediction faces three challenges: multiple foundation models (FMs) with complementary strengths per modality, pervasive missing modalities at training and test…
Unified Spatiotemporal Token Compression for Video-LLMs at Ultra-Low Retention
Junhao Du, Jialong Xue, Anqi Li +2
Video large language models (Video-LLMs) face high computational costs due to large volumes of visual tokens. Existing token compression methods typically adopt a two-stage spatiot…
Towards Trustworthy Multimodal Moderation via Policy-Aligned Reasoning and Hierarchical Labeling
Anqi Li, Wenwei Jin, Jintao Tong +3
Social platforms have revolutionized information sharing, but also accelerated the dissemination of harmful and policy-violating content. To ensure safety and compliance at scale,…
FlowCut: Rethinking Redundancy via Information Flow for Efficient Vision-Language Models
Jintao Tong, Wenwei Jin, Pengda Qin +5
Large vision-language models (LVLMs) excel at multimodal understanding but suffer from high computational costs due to redundant vision tokens. Existing pruning methods typically r…