331

arXiv:2512.14187v1 Announce Type: new
Abstract: Task-based measures of image quality (IQ) are critical for evaluating medical imaging systems, which must account for randomness including anatomical variability. Stochastic object models (SOMs) provide a statistical description of such variability, b…
333

arXiv:2504.15610v3 Announce Type: replace
Abstract: The current study describes a cost-effective method for adapting large language models (LLMs) for academic advising with study-abroad contexts in mind and for application in low-resource methods for acculturation. With the Mistral-7B-Instruct mode…
308

arXiv:2512.14560v1 Announce Type: new
Abstract: Image retrieval-based cross-view geo-localization (IRCVGL) aims to match images captured from significantly different viewpoints, such as satellite and street-level images. Existing methods predominantly rely on learning robust global representations …
331

arXiv:2512.13573v2 Announce Type: replace
Abstract: The ability to perform multi-modal multi-hop reasoning by iteratively integrating information across various modalities and external knowledge is critical for addressing complex real-world challenges. However, existing Multi-modal Large Language M…
339

Top European officials including President Volodymyr Zelensky agreed Tuesday to set up an international body to decide on tens of billions of euros of eventual reparations to compensate Ukraine over Russia's invasion. The International Claims Commission for Ukraine, signed by 35 countries, will asse…
310

Open letter to Reform UK leader expresses ‘dismay and anger’ at his response to racism and antisemitism allegationsNigel Farage has been told to apologise for his alleged teenage racism by 26 school contemporaries who have written an open letter telling of their “dismay and anger” at his response in…
421

arXiv:2512.14617v1 Announce Type: new
Abstract: Many practical decision-making problems involve tasks whose success depends on the entire system history, rather than on achieving a state with desired properties. Markovian Reinforcement Learning (RL) approaches are not suitable for such tasks, while…
234

arXiv:2512.13921v1 Announce Type: new
Abstract: Multi-hybrid architectures are poised to take over language modeling due to better quality and performance. We introduce a hierarchical decomposition framework for linear recurrences that allows us to develop algorithms aligned with GPU memory hierarc…
224

arXiv:2512.14596v1 Announce Type: new
Abstract: The convergence behavior of classical iterative solvers for parametric partial differential equations (PDEs) is often highly sensitive to the domain and specific discretization of PDEs. Previously, we introduced hybrid solvers by combining the classic…
321

arXiv:2512.14576v1 Announce Type: new
Abstract: This tutorial (https://tum-nlp.github.io/low-resource-tutorial) is designed for NLP practitioners, researchers, and developers working with multilingual and low-resource languages who seek to create more equitable and socially impactful language techn…
220

arXiv:2512.14442v1 Announce Type: new
Abstract: Affordance prediction, which identifies interaction regions on objects based on language instructions, is critical for embodied AI. Prevailing end-to-end models couple high-level reasoning and low-level grounding into a single monolithic pipeline and …
341

arXiv:2512.04013v2 Announce Type: replace
Abstract: As augmented large language models (LLMs) with external tools become increasingly popular in web applications, improving augmented LLM inference serving efficiency and optimizing service-level objectives (SLOs) are critical for enhancing user expe…
219

Taiwanese actor Chang Chen has been elected chair of the Taipei Golden Horse Film Festival Executive Committee, taking over from cinematographer Mark Lee Ping-bing, who completes his four-year tenure. The position carries a two-year term with the possibility of one re-election. Lee, who assumed the …
228

arXiv:2512.14309v1 Announce Type: new
Abstract: Self-supervised Learning (SSL) has become a powerful paradigm for representation learning without manual annotations. However, most existing frameworks focus on global alignment and struggle to capture the hierarchical, multi-scale lesion patterns cha…
314

arXiv:2512.14440v1 Announce Type: new
Abstract: In recent years, the state-of-the-art in unsupervised video instance segmentation has heavily relied on synthetic video data, generated from object-centric image datasets such as ImageNet. However, video synthesis by artificially shifting and scaling …
217

University College London has been awarded more than £44,000 through the National Cup Recycling Scheme’s Beyond the Bin fund to investigate why paper cups continue to be missed by mainstream recycling systems. University College London (UCL) has been named the winner of the National Cup Recycling Sc…
222

The Salvation Army Trading Company has suspended the collection of donated textiles from household waste and recycling centres, citing rising donation volumes, increasing costs and wider economic pressures affecting the used clothing sector. The Salvation Army Trading Company (SATCoL) has suspended …
224

In a year when the Göteborg Film Festival will highlight the importance of truth, it will give its international Honorary Dragon Award to a director who has been doing that all her long, storied and multi-prized career: Poland’s Agnieszka Holland.  One of European cinema’s greats, Holland will prese…
230

arXiv:2512.14061v1 Announce Type: new
Abstract: Recent diffusion-based one-step methods have shown remarkable progress in the field of image super-resolution, yet they remain constrained by three critical limitations: (1) inferior fidelity performance caused by the information loss from compression…
229

arXiv:2512.13726v1 Announce Type: new
Abstract: Unlike traditional recommendation tasks, finite user time budgets introduce a critical resource constraint, requiring the recommender system to balance item relevance and evaluation cost. For example, in a mobile shopping interface, users interact wit…
218

arXiv:2512.14541v1 Announce Type: cross
Abstract: Cloud quantum platforms give users access to many backends with different qubit technologies, coupling layouts, and noise levels. The execution of a circuit, however, depends on internal allocation and routing policies that are not observable to the…
208

arXiv:2512.14557v1 Announce Type: new
Abstract: Causal inference plays a crucial role in scientific research across multiple disciplines. Estimating causal effects, particularly the average treatment effect (ATE), from observational data has garnered significant attention. However, computing the AT…
299

Audit finds on average adults wait twice as long as children for assessment and more than 10 times as long to be treatedAdults with eating disorders in England are waiting up to 700 days for vital treatment, according to a report.The stark figures were revealed in the first report of the National Au…
233

Members of Franz Ferdinand, Del Amitri and more call for pause to schedule shake-up aimed at foregrounding ‘mainstream, easy-listening tracks’Musicians including Eddi Reader, Nadine Shah, Hamish Hawk and members of Franz Ferninand, Del Amitri, Idlewild and more have voiced disapproval at planned cha…
220

arXiv:2507.00629v3 Announce Type: replace-cross
Abstract: Understanding the generalization abilities of neural networks for simple input-output distributions is crucial to account for their learning performance on real datasets. The classical teacher-student setting, where a network is trained from…
221

Emmanuel Macron is in Marseille a month after the assassination of the brother of a young anti-drugs activist, a suspected intimidation tactic. In 2021, the French president had pledged significant resources to help France's second largest city modernise and curb the drug crime afflicting it. Now, h…
222

arXiv:2512.14098v1 Announce Type: new
Abstract: We present Cornserve, an efficient online serving system for an emerging class of multimodal models called Any-to-Any models. Any-to-Any models accept combinations of text and multimodal data (e.g., image, video, audio) as input and also generate comb…
213

arXiv:2512.14699v1 Announce Type: new
Abstract: The core challenge for streaming video generation is maintaining the content consistency in long context, which poses high requirement for the memory design. Most existing solutions maintain the memory by compressing historical frames with predefined …
221

arXiv:2504.02195v3 Announce Type: replace
Abstract: Modern recommendation systems fuse user behavior graphs and review texts but often encounter a "Fusion Gap" caused by False Negatives, Popularity Bias, and Signal Ambiguity. We propose SymCERE (Symmetric NCE), a contrastive learning framework brid…
220

arXiv:2510.09767v2 Announce Type: replace
Abstract: Graph Transformers have recently achieved remarkable progress in graph representation learning by capturing long-range dependencies through self-attention. However, their quadratic computational complexity and inability to effectively model hetero…
220

arXiv:2512.13739v1 Announce Type: new
Abstract: Artificial Intelligence Generated Content (AIGC) assisting image production triggers controversy in journalism while attracting attention from media agencies. Key issues involve misinformation, authenticity, semantic fidelity, and interpretability. Mo…
222

Once labeled “uninvestable,” China’s markets have rebounded this year. Stocks, onshore bonds, and the yuan are trending higher, signaling renewed investor confidence. Still, investors say a boost in economic growth is needed to keep up the momentum as the equity rally starts to lose steam and the pr…
210

One month after the stunning assassination of the brother of an anti-gangland activist, French president Emmanuel Macron returning to the Mediterranean port for an update on the government’s xx-euro plan to win back the streets and offer hope to fed-up citizens of France’s second largest city. The w…
209

Southwest Research Institute's novel Compact Dual Ion Composition Experiment (CoDICE) instrument aboard NASA's Interstellar Mapping and Acceleration Probe (IMAP) spacecraft has successfully collected first-light data. IMAP launched in September to help researchers better understand the boundary of t…
230

Growth in the US economy – and the president’s political survival – rest on AI. The EU must use its leverage and stand up to him The unthinkable has happened. The US is Europe’s adversary. The stark, profound betrayal contained in the Trump administration’s national security strategy should stop any…
220

Scientists have digitally reconstructed the face of a 1.5-million-year-old Homo erectus fossil from Ethiopia, uncovering an unexpectedly primitive appearance. While its braincase fits with classic Homo erectus, the face and teeth resemble much older human ancestors. This discovery challenges long-he…
220

Nepenthes khasiana oozes an enticing liquid on the rim of its pitchers that tempts its prey into a deadly trapA carnivorous pitcher plant has recently been found to use a chemical nerve agent to drug its prey and lead them to a deadly end, being consumed in digestive juices at the bottom of the pitc…
218

Government efforts to recruit examiners fail again with average test slot wait now at 22 weeks, report saysPlans to cut the wait for a driving test to seven weeks by the end of the year will not be achieved until November 2027, an audit report has found.Only a net 83 more driving test examiners have…
231

arXiv:2508.05223v2 Announce Type: replace
Abstract: The discourse about sustainable technology has emerged from the acknowledgment of the environmental collapse we are facing. In this paper, we argue that addressing this crisis requires more than the development of sustainable alternatives to curre…
232

arXiv:2509.23800v2 Announce Type: replace-cross
Abstract: Modern generative AI models like diffusion and flow matching can sample from rich data distributions, but many downstream tasks - such as experimental design or creative content generation - require a higher level of control than unconstrain…
233

arXiv:2312.04960v5 Announce Type: replace
Abstract: Vision Transformers (ViTs) have emerged as a fundamental architecture and serve as the backbone of modern vision-language models. Despite their impressive performance, ViTs exhibit notable vulnerability to evasion attacks, necessitating the develo…
220

arXiv:2512.13699v1 Announce Type: new
Abstract: This study investigates ``us versus them'' bias, as described by Social Identity Theory, in large language models (LLMs) under both default and persona-conditioned settings across multiple architectures (GPT-4.1, DeepSeek-3.1, Gemma-2.0, Grok-3.0, and…
111

In today’s newsletter: The BBC is bracing for a high-stakes legal fight after the US president sues over an edited Panorama clip, and questions mount about the broadcaster’s future funding and political vulnerabilityGood morning. On Tuesday, the UK woke to the news Donald Trump had filed his threate…
122

arXiv:2504.08349v5 Announce Type: replace
Abstract: Linear logic (LL) is a resource-aware, abstract logic programming language that refines both classical and intuitionistic logic. Linear logic semantics is typically presented in one of two ways: by associating each formula with the set of all cont…
121

arXiv:2409.19176v4 Announce Type: replace
Abstract: Awodey, later with Newstead, showed how polynomial functors with extra structure (termed ``natural models'') hold within them the categorical semantics for dependent type theory. Their work presented these ideas clearly but ultimately led them out…
109

A brilliant history of psychiatric ideas suggests we are on the cusp of a transformation in our understanding of severe mental illnessIn 1973, an American psychologist called David Rosenhan published the results of a bold experiment. He’d arranged for eight “pseudo-patients” to attend appointments a…
111

arXiv:2512.13729v1 Announce Type: new
Abstract: Various weather modelling problems (e.g., weather forecasting, optimizing turbine placements, etc.) require ample access to high-resolution, highly accurate wind data. Acquiring such high-resolution wind data, however, remains a challenging and expens…
117

The 2025 appeal is helping charities that bring divided communities together and promote toleranceThe Guardian’s Hope appeal has raised more than £350,000 for inspirational grassroots charities that bring divided communities together, promote tolerance and positive change, and tackle racism and hat…
118

arXiv:2511.15656v3 Announce Type: replace
Abstract: Large community science platforms such as iNaturalist contain hundreds of millions of biodiversity images that often capture ecological context on behaviors, interactions, phenology, and habitat. Yet most ecological workflows rely on metadata filt…
111

arXiv:2512.13710v1 Announce Type: new
Abstract: Flooding is one of the most destructive natural hazards worldwide, posing serious risks to ecosystems, infrastructure, and human livelihoods. This study combines Synthetic Aperture Radar (SAR) imagery with environmental and hydrological data to model …
111

arXiv:2512.14019v1 Announce Type: new
Abstract: Cancer progression arises from interactions across multiple biological layers, especially beyond morphological and across molecular layers that remain invisible to image-only models. To capture this broader biological landscape, we present EXAONE Path…
114

arXiv:2512.14349v1 Announce Type: new
Abstract: We present a novel geometric port-Hamiltonian formulation of redundant manipulators performing a differential kinematic task $\eta=J(q)\dot{q}$, where $q$ is a point on the configuration manifold, $\eta$ is a velocity-like task space variable, and $J(…