Knowledge catalogue
arXiv:2603.13331v2 Announce Type: replace-cross Abstract: Grokking -- the sudden generalisation that appears long after a model has perfectly memorised its training data -- has been widely observed bu
arXiv:2605.01311v1 Announce Type: new Abstract: Offline evaluation of language models from usage logs is biased when model choice is confounded: the same user-side factors that influence which model i
arXiv:2411.04696v5 Announce Type: replace Abstract: Learning correlations from data forms the foundation of today's machine learning (ML) and artificial intelligence research. While contemporary metho
arXiv:2605.01407v1 Announce Type: cross Abstract: Masked Language Modeling (MLM) pre-training is one of the primary ways to initialize Neural Information Retrieval (IR) models prior to retrieval fine-
arXiv:2602.13595v2 Announce Type: replace Abstract: Neural scaling laws provide a predictable recipe for AI advancement: reducing numerical precision should linearly improve computational efficiency a
arXiv:2605.01704v1 Announce Type: new Abstract: When copies of the same language model are prompted to debate, they produce diverse phrasings of one perspective rather than diverse perspectives. Multi
arXiv:2604.04465v2 Announce Type: replace Abstract: This paper identifies a structural limitation in current multimodal AI architectures that is topological rather than parametric. Contrastive alignme
arXiv:2602.12361v2 Announce Type: replace Abstract: Human-machine interfaces in industrial automation need sensing modules that monitor operator actions and physiological state. This is important in f
arXiv:2504.15077v5 Announce Type: replace Abstract: Large Language Models (LLMs) can translate natural language into SQL, but small models struggle with multi-table and complex queries in Zero-Shot Le
arXiv:2605.02496v1 Announce Type: cross Abstract: Tibetan text-to-speech (TTS) has long been challenged by scarce speech resources, significant dialectal variation, and the complex mapping between wri
arXiv:2605.02041v1 Announce Type: new Abstract: The extraction of entities and relationships from threat intelligence reports into structured formats, such as cybersecurity knowledge graphs, is essent
arXiv:2507.15774v2 Announce Type: replace Abstract: While deep learning is facing an homogenization across modalities led by Transformers, they are still challenged by shallow linear models in the tim
arXiv:2601.05300v2 Announce Type: replace-cross Abstract: Reasoning-oriented language models typically expose explicit reasoning as a long, front-loaded chain of 'thinking' tokens before the main outp
arXiv:2603.07119v2 Announce Type: replace Abstract: Recent text-to-image models have improved global realism, but text rendering remains a persistent failure mode: images may look convincing overall,
arXiv:2605.00737v1 Announce Type: new Abstract: Agentic AI architectures augment LLMs with external tools, unlocking strong capabilities. However, tool use is not always beneficial; some calls may be
arXiv:2605.01201v1 Announce Type: new Abstract: Task success has historically been the primary measure of policy performance in imitation learning (IL) research. This characteristics strictly limits t
arXiv:2605.02767v1 Announce Type: new Abstract: Diffusion models have recently demonstrated strong performance for image restoration tasks, including super-resolution. However, their large model size
arXiv:2602.11083v2 Announce Type: replace Abstract: Remote change detection in LLMs is a difficult problem. Existing methods are either too expensive for deployment at scale, or require initial white-
arXiv:2510.15545v4 Announce Type: replace Abstract: Accelerating the inference of large language models (LLMs) has been a critical challenge in generative AI. Speculative decoding (SD) substantially i
arXiv:2605.01110v1 Announce Type: new Abstract: Graph neural tangent kernels give a principled infinite-width theory for graph neural networks, but inherit a basic limitation of graph models: they see
arXiv:2605.01364v1 Announce Type: new Abstract: The building energy community lacks a foundational thermal model, i.e., a single pretrained model capable of generalizing across diverse buildings, clim
arXiv:2605.00972v1 Announce Type: cross Abstract: Earth system science is producing increasingly large, high-dimensional datasets from physics based Earth system models to AI-based weather and climate
arXiv:2603.26013v2 Announce Type: replace Abstract: Multilingual NLP is often treated as a route to global inclusion, but linguistic coverage and cultural competence frequently diverge. This paper syn
arXiv:2605.02223v1 Announce Type: cross Abstract: Recent advances in voice cloning and text-to-speech synthesis have made partial speech manipulation - where an adversary replaces a few words within a
arXiv:2605.01705v1 Announce Type: cross Abstract: Jamming attacks are proliferating and pose a significant threat to the security of 5G and beyond networks. These attacks target 5G radio frequency (RF
arXiv:2605.01663v1 Announce Type: new Abstract: We propose Flow-Anchored Noise-conditioned Q-Learning (FAN), a highly efficient and high-performing offline reinforcement learning (RL) algorithm. Recen
arXiv:2605.00883v1 Announce Type: new Abstract: Face swapping has witnessed significant progress in recent years, largely driven by advances in deep generative models such as GANs and diffusion models
arXiv:2605.00721v1 Announce Type: cross Abstract: The Room Acoustics and Speaker Distance Estimation (SDE) Challenge at ICASSP 2025 explores the effectiveness of augmented room impulse response (RIR)
arXiv:2507.04277v2 Announce Type: replace Abstract: Real-time low-light image enhancement on mobile and embedded devices requires models that balance visual quality and computational efficiency. Exist
arXiv:2605.02026v1 Announce Type: new Abstract: AC Optimal Power Flow (ACOPF) and Security-Constrained Unit Commitment (SCUC) are fundamental optimization problems in power system operations. ACOPF se
arXiv:2605.01498v1 Announce Type: new Abstract: Visual query localization (VQL) aims to predict the spatio-temporal response of the most recent occurrence in a sequence given a query. Currently, most
arXiv:2605.02615v1 Announce Type: cross Abstract: The lack of analytical models describing diffusion time dependence at intermediate time scales in complex tissue microstructure limits the accurate qu
arXiv:2603.20999v2 Announce Type: replace-cross Abstract: Adaptive 360{eg} video streaming for teleoperation faces two coupled challenges: viewport prediction under uncertain gaze patterns and bitrate
arXiv:2510.05950v2 Announce Type: replace Abstract: Time series classification (TSC) spans diverse application scenarios, yet labeled data are often scarce, making task-specific training costly and in
arXiv:2605.01928v1 Announce Type: new Abstract: Neural networks increasingly embed non-differentiable components (spiking neurons, quantized layers, discrete routing, blackbox simulators, etc.) where
arXiv:2605.01700v1 Announce Type: new Abstract: Existing zero-shot Object Goal Navigation (ObjectNav) methods often exploit commonsense knowledge from large language or vision-language models to guide
arXiv:2605.01761v1 Announce Type: new Abstract: Text-to-Video (T2V) models have demonstrated remarkable capability in generating temporally coherent videos from natural language prompts, yet they also
arXiv:2605.00895v1 Announce Type: cross Abstract: The second-order harmonic (2f) component generated by twin-rotary compressor is a dominant low-frequency noise source of variable refrigerant flow (VR
arXiv:2605.01950v1 Announce Type: new Abstract: World models enable long-horizon planning by internally generating and evaluating imagined trajectories, making them a promising foundation for generali
arXiv:2605.02193v1 Announce Type: cross Abstract: We give new examples of graphs and trees with dominating set sequences that are not log-concave. These examples were generated by PatternBoost, a tran
arXiv:2506.20588v2 Announce Type: replace Abstract: The increasing ubiquity of video content and the corresponding demand for efficient access to meaningful information have elevated video summarizati
arXiv:2605.01659v1 Announce Type: new Abstract: The rapid growth of video content across domains such as surveillance, education, and social media has made efficient content understanding increasingly
arXiv:2605.00907v1 Announce Type: new Abstract: Large language models (LLMs) and multimodal large models (MLLMs) are increasingly used for transportation tasks such as regulation question answering, t
arXiv:2605.02743v1 Announce Type: cross Abstract: The field of sensor-based human activity recognition (HAR) mainly uses posture, motion and context data of Inertial Measurement Units (IMUs) to identi
arXiv:2605.02853v1 Announce Type: new Abstract: Understanding whether deep neural networks are effectively optimized remains challenging, as training occurs in highly nonconvex landscapes and standard
arXiv:2605.01234v1 Announce Type: new Abstract: We present TT4D, a large-scale, high-fidelity table tennis dataset. It provides 140+ hours of reconstructed singles and doubles gameplay from monocular
arXiv:2605.00224v1 Announce Type: new Abstract: Aligning large language models (LLMs) with human preferences is commonly done via reinforcement learning from human feedback (RLHF) with Proximal Policy
arXiv:2510.04142v2 Announce Type: replace Abstract: This paper identifies a critical yet underexplored challenge in reasoning alignment from multiple multi-modal large language models (MLLMs): In non-
arXiv:2602.07262v3 Announce Type: replace Abstract: Second-order feature statistics are central to texture recognition, yet existing mechanisms exhibit a structural tension: bilinear pooling and Gram
arXiv:2605.01512v1 Announce Type: new Abstract: Grounding traffic accidents in real CCTV footage is a rare-event problem where training on labeled accident video is often prohibited, yet accurate join
arXiv:2605.02765v1 Announce Type: cross Abstract: LLMs are increasingly used for end-user task planning, yet their black-box nature limits users' ability to ensure reliability and control. While recen
arXiv:2602.07784v3 Announce Type: replace Abstract: Camera-based adaptive traffic signal control is inherently partially observable: detections can be missed, vehicle speeds and distances can be noisy
arXiv:2602.02056v2 Announce Type: replace-cross Abstract: Ultrafast online learning is essential for high-frequency systems, such as controls for quantum computing and nuclear fusion, where adaptation
arXiv:2605.02126v1 Announce Type: new Abstract: Ultrasound foundation models have achieved strong performance on structured prediction tasks but remain exclusively vision-based, limiting zero-shot and
arXiv:2605.00315v1 Announce Type: cross Abstract: As climate extreme and disaster events become more frequent and intense, Geospatial Artificial Intelligence (GeoAI) has emerged as a transformative ap
arXiv:2208.01899v2 Announce Type: replace Abstract: Imitation learning learns a policy from expert trajectories. While the expert data is believed to be crucial for imitation quality, it was found tha
arXiv:2605.00842v1 Announce Type: cross Abstract: Emergent misalignment, where fine-tuning on narrow, non-harmful tasks induces harmful behaviors, poses a key challenge for AI safety in LLMs. Despite
arXiv:2605.00826v1 Announce Type: cross Abstract: Text-to-video retrieval enables users to find relevant video content using natural language queries, a task that has grown increasingly important with
arXiv:2605.02380v1 Announce Type: new Abstract: Real-time crack segmentation is vital for structural health monitoring but is plagued by aleatoric uncertainties arising from varying lighting, blur, an
arXiv:2605.02762v1 Announce Type: new Abstract: Online mapping and end-to-end (E2E) planning in autonomous driving remain largely sensor-centric, leaving rich map priors, including HD/SD vector maps,