Stay updated with the latest research and technology news
arXiv:2512.20900v2 Announce Type: replace Abstract: Evaluating startups is inherently challenging in entrepreneurial finance, where investors confront severe information asymmetry and limited quantitative data. Leveraging a...
arXiv:2512.21080v3 Announce Type: replace Abstract: Field experiments (A/B tests) are often the most credible benchmark for methods (algorithms) in societal systems, but their cost and...
arXiv:2512.22137v4 Announce Type: replace Abstract: Edge-cloud collaborative inference is becoming a practical necessity for LLM-powered edge devices: on-device models often cannot afford the required reasoning...
arXiv:2512.22467v2 Announce Type: replace Abstract: In many deployed systems (multilingual ASR, cross-hospital imaging, region-specific perception), multiple pretrained specialist models coexist. Yet, new target domains often...
arXiv:2512.22716v3 Announce Type: replace Abstract: We present a theoretical study of continual and experiential learning in large language model agents that combine episodic memory with...
arXiv:2512.23471v2 Announce Type: replace Abstract: Recent advances in large language models enable documents to be represented as dense semantic embeddings, supporting similarity-based operations over large...
arXiv:2512.23526v2 Announce Type: replace Abstract: Accurate recognition of human emotional states is critical for effective human-machine interaction. Electroencephalography (EEG) offers a reliable source for emotion...
arXiv:2601.00065v2 Announce Type: replace Abstract: The open-weight language model ecosystem is increasingly defined by model composition techniques (such as weight merging, speculative decoding, and vocabulary...
arXiv:2601.01021v2 Announce Type: replace Abstract: Stochastic differential equations (SDEs) and stochastic partial differential equations (SPDEs) are fundamental for modeling stochastic dynamics across the natural sciences...
arXiv:2601.02066v3 Announce Type: replace Abstract: Replication packages are crucial for enabling transparency, validation, and reuse in software engineering (SE) research. While artifact sharing is now...
arXiv:2601.02553v3 Announce Type: replace Abstract: To support long-term interaction in complex environments, LLM agents require memory systems that manage historical experiences. Existing approaches either retain...
arXiv:2601.04443v2 Announce Type: replace Abstract: This paper presents a large language model (LLM)-based framework that adapts and fine-tunes compact LLMs for detecting cyberattacks on transformer...
arXiv:2601.04575v2 Announce Type: replace Abstract: Behavior cloning has seen a resurgence as scaling model and data sizes demonstrate strong performance. In this work, we introduce...
arXiv:2601.04589v2 Announce Type: replace Abstract: Real-world design documents (e.g., posters) are inherently multi-layered, combining decoration, text, and images. Editing them from natural-language instructions requires fine-grained,...
arXiv:2601.05199v2 Announce Type: replace Abstract: Approximation semantics capture the observable behaviour of {\lambda}-terms, with B\"ohm Trees and Taylor Expansion standing as two central paradigms. Although...
arXiv:2601.06129v2 Announce Type: replace Abstract: How many workers displaced by automation can realistically transition to safer jobs? We answer this using a validated knowledge graph...
arXiv:2601.06753v2 Announce Type: replace Abstract: Chinese paleography, the study of ancient Chinese writing, is undergoing a computational turn powered by artificial intelligence. This position paper...
arXiv:2601.06758v2 Announce Type: replace Abstract: Feedback-rich neural architectures can regenerate earlier representations and inject temporal context, making them a natural setting for strictly local synaptic...
arXiv:2601.07090v2 Announce Type: replace Abstract: This paper presents preliminary results toward a conceptual foundation for Next Generation Grid Codes (NGGCs) based on decentralized stability and...
arXiv:2601.07118v2 Announce Type: replace Abstract: Adversarial training in reinforcement learning (RL) is challenging because perturbations cascade through trajectories and compound over time, making fixed-strength attacks...
arXiv:2601.07288v2 Announce Type: replace Abstract: Although multi-view unsupervised feature selection (MUFS) has demonstrated success in dimensionality reduction for unlabeled multi-view data, most existing methods reduce...
arXiv:2601.07362v2 Announce Type: replace Abstract: Volcanic gas emissions are key precursors of eruptive activity. Yet, obtaining accurate near-surface measurements remains hazardous and logistically challenging, motivating...
arXiv:2601.07568v2 Announce Type: replace Abstract: Diffusion large language models (dLLMs) offer capabilities beyond those of autoregressive (AR) LLMs, such as parallel decoding and random-order generation....
arXiv:2601.08360v2 Announce Type: replace Abstract: Sequential recommender systems must model long-range user behavior while operating under strict memory and latency constraints. Transformer-based approaches achieve strong...