Knowledge catalogue
arXiv:2604.14910v1 Announce Type: new Abstract: Achieving high-fidelity generation in extremely few sampling steps has long been a central goal of generative modeling. Existing approaches largely rely
arXiv:2604.14197v1 Announce Type: new Abstract: Large language model (LLM) performance depends heavily on prompt design, yet prompt construction is often described and applied inconsistently. Our purp
arXiv:2604.13777v1 Announce Type: new Abstract: Large language models (LLMs) may memorize sensitive or copyrighted content, raising significant privacy and legal concerns. While machine unlearning has
arXiv:2604.13078v1 Announce Type: new Abstract: The Ramayana is among the most influential literary traditions of South and Southeast Asia, transmitted across numerous linguistic and cultural contexts
arXiv:2604.13970v1 Announce Type: new Abstract: In diagnostic reports, experts encode complex imaging data into clinically actionable information. They describe subtle pathological findings that are m
arXiv:2510.05056v2 Announce Type: replace Abstract: As programmers write code, they often edit and retry multiple times, creating rich 'interaction traces' that reveal how they approach coding tasks a
arXiv:2604.13828v1 Announce Type: new Abstract: User simulators are essential for the scalable training and evaluation of interactive AI systems. However, existing approaches often rely on shallow use
arXiv:2511.01619v2 Announce Type: replace Abstract: ParlaSpeech is a collection of spoken parliamentary corpora currently spanning four Slavic languages - Croatian, Czech, Polish and Serbian - all tog
arXiv:2604.14128v1 Announce Type: new Abstract: Rhetorical questions are asked not to seek information but to persuade or signal stance. How large language models internally represent them remains unc
arXiv:2603.08486v2 Announce Type: replace Abstract: Multimodal large language models (MLLMs) face safety misalignment, where visual inputs enable harmful outputs. To address this, existing methods req
arXiv:2604.13610v1 Announce Type: new Abstract: In computer vision, a prevailing method for quantifying dataset bias is to train a model to distinguish between datasets. High classification accuracy i
arXiv:2604.12781v1 Announce Type: new Abstract: Recently, detecting AI-generated images produced by diffusion-based models has attracted increasing attention due to their potential threat to safety. A
arXiv:2604.12336v1 Announce Type: cross Abstract: Streaming Data-Driven Optimization (SDDO) problems arise in many applications where data arrive continuously and the optimization environment evolves
arXiv:2508.01620v3 Announce Type: replace-cross Abstract: Machine Unlearning (MU) aims to selectively erase the influence of specific data points from pretrained models. However, most existing MU meth
arXiv:2510.23538v2 Announce Type: replace Abstract: The scope of neural code intelligence is rapidly expanding beyond text-based source code to encompass the rich visual outputs that programs generate
arXiv:2604.12397v1 Announce Type: new Abstract: Standard Large Language Model (LLM) pre-training typically treats corpora as flattened token sequences, often overlooking the real-world context that hu
arXiv:2604.12493v1 Announce Type: cross Abstract: LLMs can perform seemingly planning-intensive tasks, like writing coherent stories or functioning code, without explicitly verbalizing a plan; however
arXiv:2510.03174v2 Announce Type: replace-cross Abstract: Topic modeling aims to produce interpretable topic representations and topic--document correspondences from corpora, but classical neural topi
arXiv:2601.14004v4 Announce Type: replace Abstract: Mechanistic Interpretability (MI) has emerged as a vital approach to demystify the opaque decision-making of Large Language Models (LLMs). However,
arXiv:2604.12373v1 Announce Type: new Abstract: Humans use introspection to evaluate their understanding through private internal states inaccessible to external observers. We investigate whether larg
arXiv:2604.12281v1 Announce Type: cross Abstract: Style transfer aims to render a content image with the visual characteristics of a reference style while preserving its underlying semantic layout and
arXiv:2503.21708v4 Announce Type: replace-cross Abstract: Layer normalization (LN) is an essential component of modern neural networks. While many alternative techniques have been proposed, none of th
arXiv:2604.12271v1 Announce Type: new Abstract: Multimodal attributed graphs (MAGs) combine multimodal node attributes with structured relations. However, existing methods usually perform shared messa
arXiv:2509.18127v3 Announce Type: replace-cross Abstract: Sparse autoencoders (SAEs) enable interpretability research by decomposing entangled model activations into monosemantic features. However, un
arXiv:2604.12446v1 Announce Type: cross Abstract: Text-to-image (T2I) diffusion models have achieved remarkable success in image synthesis, but their reliance on large-scale data and open ecosystems i
arXiv:2604.12289v1 Announce Type: cross Abstract: Online hate speech is associated with substantial social harms, yet it remains unclear how consistently platforms enforce hate speech policies or whet
arXiv:2604.11065v1 Announce Type: new Abstract: AI systems increasingly shape high-stakes decisions in healthcare, law, defense, and education, yet existing governance paradigms -- AI Ethics, AI Safet
arXiv:2604.10865v1 Announce Type: new Abstract: Deep Clustering (DC) has emerged as a powerful tool for tabular data analysis in real-world domains like finance and healthcare. However, most existing
arXiv:2604.11424v1 Announce Type: new Abstract: Speech Language Models (SLMs) exhibit strong semantic understanding, yet their generated speech often sounds flat and fails to convey expressive intent,
arXiv:2604.09945v1 Announce Type: cross Abstract: The rapid adoption of large vision-language models (LVLMs) in recent years has been accompanied by growing fairness concerns due to their propensity t
arXiv:2604.09665v1 Announce Type: cross Abstract: While the wide adoption of refusal training in large language models (LLMs) has showcased improvements in model safety, recent works have highlighted
arXiv:2512.03676v2 Announce Type: replace Abstract: Large language models (LLMs) can reliably distinguish grammatical from ungrammatical sentences, but how grammatical knowledge is represented within
arXiv:2604.09653v1 Announce Type: cross Abstract: Beam alignment is a key challenge in directional mmWave and THz systems, where narrow beams require accurate yet low-overhead training. Existing learn
arXiv:2510.11217v2 Announce Type: replace-cross Abstract: Retrieval-Augmented Generation (RAG) combines the language understanding and reasoning power of large language models (LLMs) with external ret
arXiv:2604.09861v1 Announce Type: new Abstract: Text-to-image diffusion models exhibit strong generative performance but remain highly sensitive to prompt formulation, often requiring extensive manual
arXiv:2604.10023v1 Announce Type: cross Abstract: With the growing availability of open-sourced adapters trained on the same diffusion backbone for diverse scenes and objects, combining these pretrain
arXiv:2604.10403v1 Announce Type: new Abstract: We address jailbreaks, backdoors, and unlearning for large language models (LLMs). Unlike prior work, which trains LLMs based on their actions when give
arXiv:2511.18322v3 Announce Type: replace-cross Abstract: Learning soft continuum robot (SCR) dynamics from video offers flexibility but existing methods lack interpretability or rely on prior assumpt
arXiv:2604.11699v1 Announce Type: cross Abstract: This work aims to improve the generalization of logic-based legal reasoning systems by integrating recent advances in NLP with legal-domain adaptive f
arXiv:2604.09555v1 Announce Type: new Abstract: Multi-criteria Analysis (MCA) is used to rank alternatives based on various criteria. Key MCA methods, such as Multiple Criteria Decision Making (MCDM)
arXiv:2604.10063v1 Announce Type: new Abstract: Social media research on mental health has focused predominantly on detecting and diagnosing conditions at the individual level. In this work, we shift
arXiv:2308.12067v3 Announce Type: replace-cross Abstract: Multimodal large language models are typically trained in two stages: first pre-training on image-text pairs, and then fine-tuning using super
arXiv:2603.20725v2 Announce Type: replace Abstract: Text-to-image generation has advanced rapidly, yet it still struggles to capture the nuanced user preferences. Existing approaches typically rely on
arXiv:2603.12277v3 Announce Type: replace-cross Abstract: Language models remain vulnerable to prompt injection attacks despite extensive safety training. We trace this failure to role confusion: mode
arXiv:2509.22055v2 Announce Type: replace Abstract: We introduce RedNote-Vibe, a dataset spanning five years (pre-LLM to July 2025) sourced from lifestyle platform RedNote (Xiaohongshu), capturing the
arXiv:2604.09839v1 Announce Type: new Abstract: Activation steering is a popular white-box control technique that modifies model activations to elicit an abstract change in output behavior. It has als
arXiv:2604.09693v1 Announce Type: cross Abstract: Falls are a major cause of injury and mortality among older adults, yet most incidents occur in private indoor environments where monitoring must bala
arXiv:2604.10645v1 Announce Type: cross Abstract: There is a pressing need for better development methods and tools to keep up with the growing demand and increasing complexity of new software systems
arXiv:2602.02343v3 Announce Type: replace-cross Abstract: Methods for controlling large language models (LLMs), including local weight fine-tuning, LoRA-based adaptation, and activation-based interven
This post by Ethan Mollick likely explores AI-generated or conceptual visualizations of battles between single-celled and multicellular organisms framed as mech combat, drawing on biological history f
arXiv:2604.09076v1 Announce Type: new Abstract: Spatial transcriptomics provides a molecularly rich description of tissue organization, enabling unsupervised discovery of tissue niches -- spatially co
arXiv:2604.09189v1 Announce Type: cross Abstract: LLMs internalize safety policies through RLHF, yet these policies are never formally specified and remain difficult to inspect. Existing benchmarks ev
arXiv:2604.09358v1 Announce Type: new Abstract: Accurate prediction of nonstationary multivariate time series remains a critical challenge in complex industrial systems such as iron ore sintering. In
arXiv:2506.09067v2 Announce Type: replace-cross Abstract: Generative medical vision-language models~(Med-VLMs) are primarily designed to generate complex textual information~(e.g., diagnostic reports)
arXiv:2604.02598v2 Announce Type: replace-cross Abstract: LLM-generated explanations can make technical content more accessible, but there is a ceiling on what they can support interactively. Because
arXiv:2604.09222v1 Announce Type: cross Abstract: Audio large language models (ALLMs) enable rich speech-text interaction, but they also introduce jailbreak vulnerabilities in the audio modality. Exis
arXiv:2604.09544v1 Announce Type: cross Abstract: Large language models (LLMs) undergo alignment training to avoid harmful behaviors, yet the resulting safeguards remain brittle: jailbreaks routinely
arXiv:2506.11552v2 Announce Type: replace-cross Abstract: Quantum error correction is crucial for protecting quantum information against decoherence. Traditional codes like the surface code require su
arXiv:2604.09024v1 Announce Type: cross Abstract: Multi-modal large language models (MLLMs) have emerged as powerful tools for analyzing Internet-scale image data, offering significant benefits but al
Maybe hot take - I’ve read a bunch of RL for image generation papers over last few months and honestly it’s been pretty disappointing. All of them are variations of GRPO and all of them are incrementa