arXiv preprints from January 1, 2026 through September 23, 2026 — 21:49:39 EST
Multi-domain image-to-image translation re quires grounding semantic differences ex pressed in natural language prompts into corresponding visual transformations, while preserving unrelated structural and seman tic content. Existing methods struggle to maintain structural integrity and provide fine grained, attribute-specific control,...
Emoticons are widely used in digital communication to convey affective intent, yet their safety implications for Large Language Models (LLMs) remain largely unexplored. In this paper, we identify emoticon semantic confusion, a vulnerability where LLMs misinterpret ASCII-based emoticons to perform unintended and even destructive...
Multilingual language models (LMs) promise broader NLP access, yet current systems deliver uneven performance across the world's languages. This survey examines why these gaps persist and whether they reflect intrinsic linguistic difficulty or modeling artifacts. We organize the literature around two questions: do linguistic...
State-of-the-art text-based image editing models often struggle to balance background preservation with semantic consistency, frequently resulting either in the synthesis of entirely new images or in outputs that fail to realize the intended edits. In contrast, scene graph-based image editing addresses this limitation by providing a...
We present SceneNAT, a single-stage masked non-autoregressive Transformer that synthesizes complete 3D indoor scenes from natural language instructions through only a few parallel decoding passes, offering improved performance and efficiency compared to prior state-of-the-art approaches. SceneNAT is trained via masked modeling over...
Schiff moments of atomic nuclei are of considerable interest to experiments searching for undiscovered new physics that breaks time-reversal symmetry. I develop a simple picture of the Schiff moment of a charge distribution, and discuss the interaction of the Schiff moment of a nucleus with the field produced by an electron in an atom.
The design of quantum states at heterointerfaces has enabled a variety of emergent phenomena. Among them, molecular intercalation superlattices have attracted attention as tunable hybrid materials, formed by inserting organic molecules into van der Waals crystals, where molecular structure and chemistry provide new degrees of freedom....
The action potential constitutes the digital component of the signaling dynamics of neurons. But the biophysical nature of the full-time course of the action potential associated with changes in membrane potential is mathematically distinct from its representation as a discrete set of events that encode when action potentials are...
With the development of virtualization and AI, real-time facial avatar animation is widely used in entertainment, office, business and other fields. Against this background, blendshapes have become a common industry animation solution because of their relative simplicity and ease of interpretation. Aiming for real-time performance and...
Machine unlearning enables data holders to remove the contribution of their specified samples from trained models to protect their privacy. However, it is paradoxical that most unlearning methods require the unlearning requesters to firstly upload their data to the server as a prerequisite for unlearning. These methods are infeasible...
Information metasurfaces have emerged as pivotal components in next-generation electronic systems, with significant progress in their applications to communication, radar, and sensing. However, the current researches are mainly focused on their physical structures and system functions, while radio-frequency (RF) signal processing and...
Large Language Models (LLMs) have become indispensable across various domains, but this comes at the cost of substantial computational and memory resources. Model pruning addresses this by removing redundant components from models. In particular, block pruning can achieve significant compression and inference acceleration. However,...
The broad-line region (BLR) of active galactic nuclei (AGN) is an essential component, yet its small size keeps its origin, structure, and kinematics uncertain. Infrared interferometry with VLTI/GRAVITY is now resolving BLR-scale emission, with data for NGC 3783 consistent with a rotating, geometrically thick configuration. However,...
Dictionary learning (DL) is a core tool in signal processing and machine learning for discovering sparse representations of data. In contrast with classical successes, there is currently no practical quantum dictionary learning algorithm. We argue that this absence stems from structural mismatches between classical DL formulations and...
Glass surfaces create complex interactions of reflected and transmitted light, making single-image reflection removal (SIRR) challenging. Existing datasets suffer from limited physical realism in synthetic data or insufficient scale in real captures. We introduce a synthetic dataset generation framework that path-traces 3D glass...
Group-Relative Policy Optimization (GRPO) has emerged as an efficient paradigm for aligning Large Language Models (LLMs), yet its efficacy is primarily confined to domains with verifiable ground truths. Extending GRPO to open-domain settings remains a critical challenge, as unconstrained generation entails multi-faceted and often...
Defect related disorder during InN growth is a major challenge for making high performance electronic and optoelectronic devices. This is partly because film quality is often described using reactor specific settings instead of general physical variables. In this study, we show that plasma assisted MOCVD growth of InN can be described...
Large language model (LLM) routing assigns each query to the most suitable model from an ensemble. We introduce LLMRouterBench, a large-scale benchmark and unified framework for LLM routing. It comprises over 400K instances from 21 datasets and 33 models. Moreover, it provides comprehensive metrics for both performance-oriented...
The plasma discharges in ADITYA Upgrade Tokamak are produced by means of transformer action, in which Ohmically created plasma is driven by means of a secondary loop voltage. Due to reduction of plasma resistivity after a certain level of plasma temperature, Ohmic heating becomes poor and further achievement of temperature needs other...
In the 21st century, the era of globalization, consumers are dispersed across the globe, and brands compete for their attention and loyalty, largely within the digital realm. This reality elevates the importance of effective communication and the transmission of product value across diverse cultural contexts. This study presents a...
The paper extends three results regarding the nth root problem by embedding classes of Hilbert-space operators into the class of posinormal operators. For instance, it is shown that (i) for coposinormal operators, if T is paranormal and T^n is quasinormal, then T is normal, and (ii) for posinormal operators, if T is k-quasiparanormal...
Slogans are concise and memorable catchphrases that play a crucial role in advertising by conveying brand identity and shaping public perception. However, advertising fatigue reduces the effectiveness of repeated slogans, creating a growing demand for novel, creative, and insightful slogan generation. While recent work leverages large...
Reduced-rank regression estimates regression coefficients by imposing a low-rank constraint on the matrix of regression coefficients, thereby accounting for correlations among response variables. To further improve predictive accuracy and model interpretability, several regularized reduced-rank regression methods have been proposed....
Deep protein structure predictors such as AlphaFold provide confidence estimates (e.g., pLDDT) that are often miscalibrated and degrade under distribution shifts across experimental modalities, temporal changes, and intrinsically disordered regions. We introduce CalPro, a prior-aware evidential-conformal framework for shift-robust...
The inherent safety alignment of Large Language Models (LLMs) is prone to erosion during fine-tuning, even when using seemingly innocuous datasets. While existing defenses attempt to mitigate this via data selection, they typically rely on heuristic, instance-level assessments that neglect the global geometry of the data distribution...