Compute·~10Y2025-12-06
"287,872 Supermassive Black Holes Masses: Deep Learning Approaching Reverberation Mapping Accuracy"
"We present a population-scale catalogue of 287,872 supermassive black hole masses with high accuracy."
Signals
Research updates tracked across the four pillars.
Compute·~10Y2025-12-06
"Are LLMs Truly Multilingual? Exploring Zero-Shot Multilingual Capability of LLMs for Information Retrieval: An Italian Healthcare Use Case"
"Our detailed experimental campaign on comorbidity extraction from EHR reveals that some LLMs struggle in zero-shot, on-premises settings, and others show significant variation in performance, struggling to generalize across various diseases when compared to native pattern matching and manual annotations."
Compute·NOW2025-12-06
"Are Your Agents Upward Deceivers?"
"Large Language Model (LLM)-based agents are increasingly used as autonomous subordinates that carry out tasks for users."
Compute·~2Y2025-12-06
"Balanced Few-Shot Episodic Learning for Accurate Retinal Disease Diagnosis"
"Automated retinal disease diagnosis is vital given the rising prevalence of conditions such as diabetic retinopathy and macular degeneration."
Compute·~10Y2025-12-06
"CARL: Critical Action Focused Reinforcement Learning for Multi-Step Agent"
"However, in such multi-step settings, the conventional group-level policy optimization algorithm becomes suboptimal because of its underlying assumption that each action holds equal contribution, which deviates significantly from reality."
Compute·~10Y2025-12-06
"Chameleon: Adaptive Adversarial Agents for Scaling-Based Visual Prompt Injection in Multimodal AI Systems"
"However, this dependency on standard preprocessing operations, specifically image downscaling, creates a significant yet often overlooked security vulnerability."
Compute·~2Y2025-12-06
Neural networks systematically converge to shared spectral subspaces regardless of initialization, task, or domain
This inherent structure has implications for model reusability, multi-task learning, and potentially reducing the carbon footprint of large-scale neural models
Compute·~10Y2025-12-06
"Constrained Control of PDE Traffic Flow via Spatial Control Barrier Functions"
"In this paper, a constrained control approach to variable speed limit (VSL) control for macroscopic partial differential equations (PDE) traffic models is developed."
Compute·~10Y2025-12-06
"Counterfactual Explanations for Power System Optimisation"
"The thereby obtained explanations allow users to identify the most important differences between the real and expected market outcomes and observe which constraints have led to the solution."
Compute·~10Y2025-12-06
"David vs. Goliath: Can Small Models Win Big with Agentic AI in Hardware Design?"
"Large Language Model(LLM) inference demands massive compute and energy, making domain-specific tasks expensive and unsustainable."
Compute·~10Y2025-12-06
"Declarative Synthesis and Multi-Objective Optimization of Stripboard Circuit Layouts Using Answer Set Programming"
"This paper presents a novel approach to automated stripboard circuit layout design using Answer Set Programming (ASP)."
Compute·~10Y2025-12-06
"Detecting Perspective Shifts in Multi-agent Systems"
"This paper introduces the Temporal Data Kernel Perspective Space (TDKPS), which jointly embeds agents across time, and proposes several novel hypothesis tests for detecting behavioral change at the agent- and group-level in black-box multi-agent systems."
Compute·~10Y2025-12-06
"DraCo: Draft as CoT for Text-to-Image Preview and Rare Concept Generation"
"To this end, we propose Draft-as-CoT (DraCo), a novel interleaved reasoning paradigm that fully leverages both textual and visual contents in CoT for better planning and verification."
Compute·~10Y2025-12-06
"Enabling Ethical AI: A case study in using Ontological Context for Justified Agentic AI Decisions"
"AI agents first propose candidate knowledge structures from diverse data sources; domain experts then validate, correct, and extend these structures, with their feedback used to improve subsequent models."
Compute·~10Y2025-12-06
"Evolutionary Architecture Search through Grammar-Based Sequence Alignment"
"Neural architecture search (NAS) in expressive search spaces is a computationally hard problem, but it also holds the potential to automatically discover completely novel and performant architectures."
Compute·~10Y2025-12-06
"From Symptoms to Systems: An Expert-Guided Approach to Understanding Risks of Generative AI for Eating Disorders"
"Existing safeguards tend to overlook subtle but clinically significant cues, leaving many risks unaddressed."
Compute·~10Y2025-12-06
"GeoPE:A Unified Geometric Positional Embedding for Structured Tensors"
"Extensive experiments on image classification, object detection, and 3D semantic segmentation demonstrate that GeoPE consistently outperforms existing 2D RoPE variants and significantly enhances shape bias, confirming its ability to capture true geometric structure."
Compute·~10Y2025-12-06
"Language Models as Semantic Teachers: Post-Training Alignment for Medical Audio Understanding"
"' To enable alignment at scale, we construct a large-scale dataset by leveraging off-the-shelf large language models to translate the rich, structured metadata accompanying existing audio recordings into coherent clinical reports."
Compute·~10Y2025-12-06
"LLMs Know More Than Words: A Genre Study with Syntax, Metaphor & Phonetics"
"To analysis whether LLMs can learn these features effectively and apply them to important nature language related tasks, we introduce a novel multilingual genre classification dataset derived from Project Gutenberg, a large-scale digital library offering free access to thousands of public domain literary works, comprising thousands of sentences per binary task (poetry vs."
Compute·~2Y2025-12-06
"Meta-Learning for Quantum Optimization via Quantum Sequence Model"
"However, finding good variational parameters remains a significant challenge due to the non-convex energy landscape, often resulting in slow convergence and poor solution quality."
Compute·~10Y2025-12-06
"Mitigating Catastrophic Forgetting in Target Language Adaptation of LLMs via Source-Shielded Updates"
"Advances understanding in this domain."
Compute·~10Y2025-12-06
"Model-Based and Sample-Efficient AI-Assisted Math Discovery in Sphere Packing"
"Even a major breakthrough in dimension $n=8$, later recognised with a Fields Medal, underscores its difficulty."
Compute·~10Y2025-12-06
"Model-Free Assessment of Simulator Fidelity via Quantile Curves"
"Simulation of complex systems originated in manufacturing and queuing applications."
Compute·~10Y2025-12-06
"ReflexFlow: Rethinking Learning Objective for Exposure Bias Alleviation in Flow Matching"
"Despite tremendous recent progress, Flow Matching methods still suffer from exposure bias due to discrepancies in training and inference."
Compute·~10Y2025-12-06
"SA-IQA: Redefining Image Quality Assessment for Spatial Aesthetics with Multi-Dimensional Rewards"
"We construct SA-BENCH, the first benchmark for spatial aesthetics, comprising 18,000 images and 50,000 precise annotations."
Compute·~10Y2025-12-06
"SEAL: Self-Evolving Agentic Learning for Conversational Question Answering over Knowledge Graphs"
"To address these limitations, we introduce SEAL, a novel two-stage semantic parsing framework grounded in self-evolving agentic learning."
Compute·NOW2025-12-06
"Semantic Soft Bootstrapping: Long Context Reasoning in LLMs without Reinforcement Learning"
"As a result, it requires significant compute resources in post-training phase."
Compute·~10Y2025-12-06
"Setting up for failure: automatic discovery of the neural mechanisms of cognitive errors"
"First, as the amount of behavioural data that can be collected in experiments is often too limited to train RNNs, we use a non-parametric generative model of behavioural responses to produce surrogate data for training RNNs."
Compute·~2Y2025-12-06
"ShadowDraw: From Any Object to Shadow-Drawing Compositional Art"
"We introduce ShadowDraw, a framework that transforms ordinary 3D objects into shadow-drawing compositional art."
Compute·~10Y2025-12-06
"Small-Signal Stability Oriented Real-Time Operation of Power Systems with a High Penetration of Inverter-Based Resources"
"First, an offline analysis of a set of operating points is performed to derive a data-driven regression-based expression that captures a damping-based stability index as a function of the operating conditions."
Compute·~10Y2025-12-06
"SoK: a Comprehensive Causality Analysis Framework for Large Language Model Security"
"The framework enables consistent experimentation and comparison across diverse causality-based attack and defense methods."
Compute·~10Y2025-12-06
"Stability-Guaranteed Dual Kalman Filtering for Electrochemical Battery State Estimation"
"Accurate and stable state estimation is critical for battery management."
Compute·~10Y2025-12-06
"STELLA: Guiding Large Language Models for Time Series Forecasting with Semantic Abstractions"
"Recent adaptations of Large Language Models (LLMs) for time series forecasting often fail to effectively enhance information for raw series, leaving LLM reasoning capabilities underutilized."
Compute·~10Y2025-12-06
"Strategic Self-Improvement for Competitive Agents in AI Labour Markets"
"This paper puts forward a groundbreaking new framework that is the first to capture the real-world economic forces that shape agentic labor markets: adverse selection, moral hazard, and reputation dynamics."
Compute·~10Y2025-12-06
"Structured Document Translation via Format Reinforcement Learning"
"To address this, we propose \textbf{Format Reinforcement Learning (FormatRL)}, which employs Group Relative Policy Optimization on top of a supervised fine-tuning model to directly optimize novel structure-aware rewards: 1) TreeSim, which measures structural similarity between predicted and reference XML trees and 2) Node-chrF, which measures translation quality at the level of XML nodes."
Compute·~10Y2025-12-06
"The AI Consumer Index (ACE)"
"We introduce the first version of the AI Consumer Index (ACE), a benchmark for assessing whether frontier AI models can perform high-value consumer tasks."
Compute·~10Y2025-12-06
"The Universal Weight Subspace Hypothesis"
"We provide the first large-scale empirical evidence that demonstrates that neural networks systematically converge to shared spectral subspaces regardless of initialization, task, or domain."
Compute·~10Y2025-12-06
"Toward Continuous Neurocognitive Monitoring: Integrating Speech AI with Relational Graph Transformers for Rare Neurological Diseases"
"Patients with rare neurological diseases report cognitive symptoms -'brain fog'- invisible to traditional tests."
Compute·~10Y2025-12-06
"TV2TV: A Unified Framework for Interleaved Language and Video Generation"
"Video generation models are rapidly advancing, but can still struggle with complex video outputs that require significant semantic branching or repeated high-level reasoning about what should happen next."