4 papers · 1 filter
Why MLLMs Struggle to Count: Overcoming Individuation and Aggregation Bottlenecks with ConvStack
Liwei Che, Yihao Quan, Sen Fang +4
Multimodal Large Language Models (MLLMs) consistently struggle with fine-grained visual counting, yet the underlying causes remain poorly understood. In this work, we present a mec…
Counting Circuits: Mechanistic Interpretability of Visual Reasoning in Large Vision-Language Models
Liwei Che, Zhiyu Xue, Yihao Quan +7
Counting serves as a simple but powerful test of a Large Vision-Language Model's (LVLM's) reasoning; it forces the model to identify each individual object and then add them all up…
Box2Flow: Instance-based Action Flow Graphs from Videos
Jiatong Li, Kalliopi Basioti, Vladimir Pavlovic
A large amount of procedural videos on the web show how to complete various tasks. These tasks can often be accomplished in different ways and step orderings, with some steps able…
CIC-BART-SSA: Controllable Image Captioning with Structured Semantic Augmentation
Kalliopi Basioti, Mohamed A. Abdelsalam, Federico Fancellu +2
Controllable Image Captioning (CIC) aims at generating natural language descriptions for an image, conditioned on information provided by end users, e.g., regions, entities or even…