AI
Defining Decentralization: An Ontological Perspective
Researchers from the University of Edinburgh have proposed an ontological framework to define decentralization in computer science. They argue that cu...
AI
AI
ELVAE: Evidential Learning-Based Variational Autoencoder for Uncertainty-Aware Generation
Researchers have proposed a new AI model called ELVAE (Evidential Learning-Based Variational Autoencoder) that can generate images with uncertainty-aw...
AI
Hamilton-Zero: A Neural Tensor-Network Foundation Model for Ground States of Arbitrary Quadratic Qubit Hamiltonians
Researchers have developed a neural tensor-network foundation model that can efficiently compute ground states of arbitrary quadratic qubit Hamiltonia...
AI
DFM Mimir v1: An Open HRM Delivering Frontier Performance at 1B Parameters Using Only Permissible Post-Training Data
Researchers have introduced Mimir v1, a large language model with 1 billion parameters. Unlike many other models, it was trained from scratch using on...
AI
Information Geometry of Message Passing
Researchers have developed a new method for message passing in variational inference called natural-gradient message passing (NGMP). This approach is ...
AI
Beyond Raw Transcripts: Structured Persona Extraction for LLM-Based Digital Twins
Researchers have developed a new approach to creating 'digital twins' - AI models that simulate human behavior. Instead of using raw transcripts or su...
AI
Building and Evaluating a Synthetic Bengali Speech Resource for Telecom Customer Care
Researchers have created a synthetic Bengali speech dataset for use in customer-facing applications like telecom customer care. The dataset contains 1...
AI
Exploratory As-Analyzed No-Detection of Culturally-Marked Predicate-Triggered PII Amplification in a Synthetic-English RAG Probe: A Predicate-Resource-Confounded Audit
Researchers from a team led by Yanhang Li have conducted an exploratory study on whether a retrieval-augmented generation (RAG) system leaks more pers...
AI
Self-Speculation for Faster Reasoning Models
Researchers have developed a new method called Self-Speculation for Reasoning Models (SSR) to speed up large language models' reasoning processes. SSR...
AI
TriPLU: Bypassing the Gate with Direct Trilinear Product FFNs in Tiny Language Models
Researchers have proposed a new type of feed-forward neural network called TriPLU for use in tiny language models. Instead of using the usual gated FF...
AI
Multilingual Verifier Bias in RLVR: Benchmark, Rollout Diagnosis, and the Cross-Lingual Selection Bottleneck
Researchers have found that a common technique for training large language models on mathematical reasoning tasks, called reinforcement learning with ...
AI
Research Paper Quality Recognition Through Textual Feature Analysis
Researchers have developed a method to evaluate the quality of scientific papers based on their titles and abstracts. The approach uses various text a...