Vision-language models, speech and cross-modal systems that read, look and listen in the same forward pass.
Long-video understanding remains challenging for multimodal large language models, because temporally extended videos often contain thousands of frames and are therefore expensive to process exhaustively. Existing methods usually construct compact visual inputs from long videos under a limited visual budget.
This paper presents our solution to the 2026 SoccerNet VQA Challenge. We first develop a cost-effective data synthesis pipeline driven by a Vision-Language Model (VLM), which systematically restructures raw domain data into diverse VQA samples, including concise answers and long-form responses.
Vision-language-action (VLA) models inherit a shared synchronous clock from vision-language pretraining, processing every input at one rate. This is misaligned with physical interaction, where a high-frequency modality changes at hundreds of hertz, vision evolves more slowly, and language stays constant across an episode.
Automating compliance check for geometry-intensive regulations remains a significant technical bottleneck in Building Information Modeling (BIM), primarily due to the semantic disparity between high-level regulatory logic and structured IFC data. Existing methods, often reliant on static rule templates, struggle to traverse multi-hop reasoning chains or resolve latent spatial dependencies across multiple building entities.
In this paper, we address the problem of zero-shot understanding of accidents from surveillance videos by identifying when an impact event occurs, what type of impact it is, and where in the frame it occurs using natural language. We propose a three-stage pipeline that decomposes the accident understanding into when, what, and where.
We propose a multi-agent collaborative framework built upon a lightweight Multimodal Large Language Model (MLLM), specifically designed for social intelligence reasoning. A key feature of our approach is that both the training and inference phases are augmented via knowledge distillation.
arXiv:2606. 10504v1 Announce Type: new Abstract: Cross-modal knowledge distillation (CMKD) studies how a (large) teacher model trained on one type of data (e.
arXiv:2606. 10198v1 Announce Type: cross Abstract: Hallucination detection in large language and vision-language models is increasingly framed as selective prediction, where a detector assigns a confidence score and abstains when confidence is low.
arXiv:2606. 09942v1 Announce Type: cross Abstract: Microservice systems are widely used to build cloud applications, yet their complexity makes failures inevitable, degrading user experience and causing economic loss.
arXiv:2606. 10064v1 Announce Type: cross Abstract: Small-model agentic post-training is bottlenecked less by the algorithm than by the trajectory substrate it consumes.
arXiv:2606. 10738v1 Announce Type: cross Abstract: Recent multimodal large language models mainly process audio as monaural signals, thereby discarding the spatial cues contained in spatial audio for sound localization, spatial relation reasoning, and spatial scene understanding.
arXiv:2606. 09853v1 Announce Type: new Abstract: A central objective in multimodal learning is to capture synergy: task-relevant information that arises only from the joint use of multiple modalities, and is not available from any single modality alone.
arXiv:2606. 10862v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models achieve strong performance on standard manipulation benchmarks, but most evaluations assume that task-relevant objects are fully visible.
arXiv:2606. 10918v1 Announce Type: cross Abstract: The recent trend in scaling models for robot learning has resulted in impressive policies that can perform various manipulation tasks and generalize to novel scenarios.
arXiv:2606. 10180v1 Announce Type: cross Abstract: We introduce flow control of vision-language-action (VLA) models, a simple and effective way to steer VLA actions in real-time through generic inputs, such as a keyboard.
arXiv:2606. 11190v1 Announce Type: new Abstract: Cross-modal alignment (CA) and cross-modal prediction (CP) are the dominant paradigms for multimodal representation learning, yet there is no systematic understanding of when each succeeds, when each fails, and when cross-modal training helps at all -- a gap that leaves practitioners, especially in scientific domains like biomedicine or astrophysics, with heterogeneous instruments and multiple levels of organization and measurement, unable to diagnose why standard methods underperform the best single modality.
arXiv:2603. 22017v2 Announce Type: replace Abstract: This work presents a collection of multi-modal domain adapted large language models built upon the instruction tuned variants of open weight models (Gemma 3, Qwen 3, Gemma 4) using a relatively small dataset of around 50 million tokens.
arXiv:2606. 10066v1 Announce Type: cross Abstract: Medical vision-language models (VLMs) are evaluated on public benchmarks whose images and question-answer pairs have been freely downloadable for years, yet reported accuracy assumes these examples were absent from pretraining.
arXiv:2606. 10819v1 Announce Type: cross Abstract: RS-MLLMs enable natural-language understanding and spatial reasoning over earth observation imagery.
arXiv:2606. 11106v1 Announce Type: cross Abstract: A global shortage of trained sonographers limits prenatal ultrasound screening in low- and middle-income countries, where over half of pregnant women receive no skilled sonography.