Pretrained video generation models distill internet-scale motion priors into robot policies, but full video is a redundant signal for action prediction. We argue that dense 3D motion is the medium through which video pretraining best transfers to robot actions, and that predicting motion rather than video affords bette...
Jeremy A. Collins, Namra Patel, A. Agarwal et al.· 0 citations
Recent multimodal language models can reason directly about robot scenes, but it is unclear whether they need to repeatedly reinterpret raw visual observations at every step of an interaction. We propose shifting routine spatial reasoning from images to text and symbols through a 3D geometric abstraction that is refres...
On-device RGB photo enhancement is a popular task that is needed for quality improvement of photos captured with mobile devices as well as for retouching of images acquired through other sources. This creates a need for efficient deep learning models for this problem that are not only performant but can also be deploye...
Andrey Ignatov, Georgy Perevozchikov, Radu Timofte et al.· 0 citations
Generalised Additive Models (GAMs) are a cornerstone of interpretable machine learning, offering transparent predictions through structured additive decompositions. Despite advances from classical GAMs to modern neural and spline-based variants, two notable gaps remain. First, comprehensive benchmarking of spline-based...
George Dimitriadis, Damian Machlanski, Paul Clarke et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
With the increasing importance of mobile camera quality in modern smartphones, growing attention has been directed toward computational photography and advanced image signal processing (ISP) algorithms aimed at enhancing various perceptual attributes of captured images. This creates a need for efficient deep learning m...
Andrey Ignatov, Georgy Perevozchikov, Radu Timofte et al.· 0 citations
Real-world geospatial decision support requires more than single-step retrieval or code generation. Disaster-response questions often involve multi-source workflows that combine spatial databases, map-derived features, satellite imagery, raster products, and reusable geospatial artifacts, where correctness depends on p...
Muhammad Sohail Danish, Gilles Q. Hacheme, Akram Zaytar et al.· 0 citations
We investigate the automated generation of executable robot reinforcement learning (RL) policies from natural-language task descriptions. Rather than using large language models (LLMs) as execution-time decision makers, we use them as designers that construct RL pipelines before deployment, including Markov decision pr...
Hong Je-Gal, Jinwoo Kim, Chanbin Yi et al.· 0 citations
This study evaluates whether large language models (LLMs) can reliably replace or complement human research assistants in linguistically complex annotation. In a controlled annotation task, 73 human annotators (local students and crowdworkers with and without a background in linguistics) and more than 40 LLM configurat...
Knowledge distillation improves a student model's reasoning ability through supervision from a teacher. On-policy self-distillation (OPSD) uses the same model as both teacher and student under different contexts: the student generates reasoning trajectories from the question alone, while the teacher additionally receiv...
Multimodal agents use web and image tools to answer questions beyond their internal knowledge. Some search tasks use images mainly to identify an initial entity, then rely on text retrieval. These tasks offer limited insight into how agents connect information across images, objects, and search steps. We therefore focu...
ZhengXian Wu, Qianhao Nie, Zhuohong Chen et al.· 0 citations
Real-world sensing systems are inherently imperfect: sensors may fail, intermittently disconnect, operate with asynchronous sampling rates, and be subject to noise, leading to measurement streams that are irregular, incomplete, and heterogeneous. Nevertheless, sensing strategies for real-time state reconstruction typic...
Matteo Romilio Pasqual, Matteo Tomasetto, J. Nathan Kutz et al.· 0 citations
The value of memory lies not only in recall but in using it to provide appropriate assistance. Recent benchmarks test memory use in personalisation and agentic tasks, but seldom evaluate assistance without fixed answers or integration across multiple participants. Group chats make both challenges more acute: an assista...
Zhanghao Hu, Linhai Zhang, Qian Zhao et al.· 0 citations
