14 accepted papers
Diffusion Large Language Models (DLLMs) are emerging as a powerful alternative to the dominant Autoregressive Large Language Models, offering efficient parallel generation and capable global context modeling. However, the practical application of DLLMs is hindered by a critical architectural constra…
Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities, yet their ability to ground language in complex, interactive environments such as video games remains a critical frontier. Existing benchmarks are inadequate for this purpose: real-world datasets like RefCOCO introduce a
This paper presents ScaleCap, a scalable image captioning strategy that generates comprehensive and detailed image captions. The key challenges of high-quality image captioning lie in the inherent biases of LVLMs: multimodal bias resulting in imbalanced descriptive granularity, offering detailed acc…
While Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities for reasoning and self-correction at the textual level, these strengths provide minimal benefits for complex tasks centered on visual perception, such as Chart Parsing. Existing models often struggle with visually d…
Semantic scene understanding in bird-eye view (BEV) plays a crucial role in autonomous driving. A common approach to generating BEV maps from LiDAR point-cloud data involves constructing a pillar-level representation by projecting 3D point clouds onto a 2D plane. This process partially discards spat
Multi-robot cooperative tracking, as a vital sub-field of multi-robot collaboration, exhibits significant potential in areas such as military reconnaissance and emergency rescue. Conventional dynamic object tracking methods often face issues of incomplete target detection and even loss in complex sc
Despite the strong performance of ColPali/ColQwen2 in Visualized Document Retrieval (VDR), its patch-level embedding approach leads to excessive memory usage. This empirical study investigates methods to reduce patch embeddings per page while minimizing performance degradation. We evaluate two token…
Large vision-language models (LVLMs) have recently achieved rapid progress, sparking numerous studies to evaluate their multi-modal capabilities. However, we dig into current evaluation works and identify two primary issues: 1) Visual content is unnecessary for many samples. The answers can be direc…
This work develops a fast mission planning framework named planning decision tree (PDT), that can handle large-scale multi-robot systems with temporal logic specifications in real time. Specifically, PDT builds a tree incrementally to represent the task progress. The system states are modeled by bot
As a prominent parameter-efficient fine-tuning technique in NLP, prompt tuning is being explored its potential in computer vision. Typical methods for visual prompt tuning follow the sequential modeling paradigm stemming from NLP, which represents an input image as a flattened sequence of token embe…
"Modality alignment serves as the cornerstone for large multi-modal models (LMMs). However, the impact of different attributes (e.g., data type, quality, and scale) of training data on facilitating effective alignment is still under-explored. In this paper, we delve into the influence of training da…
We present the ShareGPT4Video series, aiming to facilitate the video understanding of large video-language models (LVLMs) and the video generation of text-to-video models (T2VMs) via dense and precise captions. The series comprises: 1) ShareGPT4Video, 40K GPT4V annotated dense captions of videos wit…
Learning-based path planning is becoming a promising robot navigation methodology due to its adaptability to various environments. However, the expensive computing and storage associated with networks impose significant challenges for their deployment on low-cost robots. Motivated by this practical…
With the advent of solid-state LiDAR, a series of related studies have boosted the development of Simultaneous Localization and Mapping (SLAM). However, existing methods cannot work well in indoor environments. In the letter, the reflectivity measurement of the solid-state LiDAR is exploited to impr