[Crossposted on lesswrong, see here for prior posts] The following statements seem to be both important for AI safety and are not widely agreed upon. These are my opinions, not those of my employer or colleagues. As is true for anything involving AI, there is significant uncertainty about everything written below. However, for readability, I present these points in their strongest form, without hedges and caveats. That said, it is essential not to be dogmatic, and I am […]

Ver mais

Accurate detection and localization of traumatic injuries in abdominal CT scans remains a critical challenge in emergency radiology, primarily due to severe scarcity of annotated medical data. This paper presents a label-efficient approach combining self-supervised pre-training with semi-supervised detection for 3D medical image analysis. We employ patch-based Masked Image Modeling (MIM) to pre-train a 3D U-Net encoder on 1,206 CT volumes without annotations, learning robust anatomical representations. The pretrained encoder enables two downstream clinical tasks: 3D injury detection […]

Ver mais

In this study, we developed a chatbot-based system for detecting early signs of depression and verified its effectiveness through experimental evaluations and user surveys. Em-phasizing that it does not rely on medical checklists, the system is designed to auto-matically extract three linguistic features associated with depression—frequent use of first-person pronouns, pessimistic expressions, and obsessive-compulsive writing styles—from natural user conversations. Multiple models were constructed for these features, and an ensemble layer integrates their outputs for a comprehensive judgment. The […]

Ver mais

In offline reinforcement learning (RL), single-step temporal-difference (TD) learning can suffer from bootstrapping error accumulation over long horizons. Action-chunked TD methods mitigate this by backing up over multiple steps, but can introduce suboptimality by restricting the policy class to open-loop action sequences. To resolve this trade-off, we present Chunk-Guided Q-Learning (CGQ), a single-step TD algorithm that guides a fine-grained single-step critic by regularizing it toward a chunk-based critic trained using temporally extended backups. This reduces compounding error while […]

Ver mais

arXiv:2602.23587v1 Announce Type: new Abstract: Knowledge distillation transfers large teacher models to compact student models, enabling deployment on resource-limited platforms while suffering minimal performance degradation. However, this paradigm could lead to various security risks, especially model theft. Existing defenses against model theft, such as watermarking and secure enclaves, focus primarily on identity authentication and incur significant resource costs. Aiming to provide post-theft accountability and traceability, we propose a novel fingerprinting framework that superimposes device-specific Physical Unclonable Function (PUF) […]

Ver mais

How are you, hacker? 🪐Want to know what’s trending right now?: The Techbeat by HackerNoon has got you covered with fresh content from our trending stories of the day! Set email preference here. ## SEO Isn’t Dead But Your Strategies Have to Change By @deeflect [ 10 Min read ] CTRs are down 30%, and AI Overviews are everywhere. SEO isn’t dead, but the rules shifted. Here’s what GEO is and why brand mentions now outrank backlinks 3:1 […]

Ver mais

arXiv:2601.14300v1 Announce Type: new Abstract: Hard-label black-box settings, where only top-1 predicted labels are observable, pose a fundamentally constrained yet practically important feedback model for understanding model behavior. A central challenge in this regime is whether meaningful gradient information can be recovered from such discrete responses. In this work, we develop a unified theoretical perspective showing that a wide range of existing sign-flipping hard-label attacks can be interpreted as implicitly approximating the sign of the true loss gradient. […]

Ver mais

arXiv:2603.07014v1 Announce Type: cross Abstract: Regression with distribution-valued responses and Euclidean predictors has gained increasing scientific relevance. While methodology for univariate distributional data has advanced rapidly in recent years, multivariate distributions, which additionally encode dependence across univariate marginals, have received less attention and pose computational and statistical challenges. In this work, we address these challenges with a new regression approach for multivariate distributional responses, in which distributions are modeled within the semiparametric nonparanormal family. By incorporating the nonparanormal […]

Ver mais

Previously, businesses have taken disruption lightly. Something to live with, something to respond to, something that will culminate. This closure of factories, collapse of suppliers, logistical difficulties, and system failures are all unwanted guests, and they push leaders to enter the crisis mode. The fag-end work is in teams, and decisions are fast; spending is too big, and hopefully it will be the environment, the next shock to be felt somewhere in the environment. Such a way of […]

Ver mais