Learning Adaptive Parallel Reasoning with Language Models
Scaling inference-time computation has substantially improved the reasoning capabilities of language models. However, existing methods have significant limitations: serialized chain-of-thought approaches generate overly long outputs, leading to increased latency and exhausted context windows, while parallel methods such as self-consistency suffer from insufficient...
Long-lived entanglement of a spin-qubit register in silicon photonics
Color centers provide an optical interface to quantum registers based on electron and nuclear spin qubits in solids. The T center in silicon is an emerging spin-photon interface that combines telecom O-band optical transitions and a long-lived electron spin in a scalable...
Generating Dark Matter Subhalo Populations Using Normalizing Flows
Strong gravitational lensing is a powerful tool for probing the nature of dark matter, as lensing signals are sensitive to the dark matter substructure within the lensing galaxy. We present a comparative analysis of strong gravitational lensing signatures generated by dark matter...
Trends in Frontier AI Model Count: A Forecast to 2028
Governments are starting to impose requirements on AI models based on how much compute was used to train them. For example, the EU AI Act imposes requirements on providers of general-purpose AI with systemic risk, which includes systems trained using greater than...
Manifold Induced Biases for Zero-shot and Few-shot Detection of Generated Images
Distinguishing between real and AI-generated images, commonly referred to as ‘image detection’, presents a timely and significant challenge. Despite extensive research in the (semi-)supervised regime, zero-shot and few-shot solutions have only recently emerged as promising alternatives. Their main advantage is in alleviating...
LAPP: Large Language Model Feedback for Preference-Driven Reinforcement Learning
We introduce Large Language Model-Assisted Preference Prediction (LAPP), a novel framework for robot learning that enables efficient, customizable, and expressive behavior acquisition with minimum human effort. Unlike prior approaches that rely heavily on reward engineering, human demonstrations, motion capture, or expensive pairwise...
Agent for User: Testing Multi-User Interactive Features in TikTok
TikTok, a widely-used social media app boasting over a billion monthly active users, requires effective app quality assurance for its intricate features. Feature testing is crucial in achieving this goal. However, the multi-user interactive features within the app, such as live streaming,...
Unifying Image Counterfactuals and Feature Attributions with Latent-Space Adversarial Attacks
Counterfactuals are a popular framework for interpreting machine learning predictions. These what if explanations are notoriously challenging to create for computer vision models: standard gradient-based methods are prone to produce adversarial examples, in which imperceptible modifications to image pixels provoke large changes...
Under Pressure: Contextualizing Workplace Stress Towards User-Centered Interventions
Stress is a pervasive challenge that significantly impacts worker health and well-being. Workplace stress is driven by various factors, ranging from organizational changes to poor workplace design. Although individual stress management strategies have been shown to be effective, current interventions often overlook...
A Sensor-Driven Optimization Framework for Asset Management in Energy Systems: Implications for Full and Partial Digital Transformation in Hydro Fleets
This paper proposes a novel prognostics-driven approach to optimize operations and maintenance (O&M) decisions in hydropower systems. Our approach harnesses the insights from sensor data to accurately predict the remaining lifetime distribution of critical generation assets in hydropower systems, i.e., thrust bearings,...