AI/ML arXiv cs.AI

Bring My Cup! Personalizing Vision-Language-Action Models with Visual Attentive Prompting

Introduces Visual Attentive Prompting (VAP), a training-free adapter that enables Vision-Language-Action models to identify and manipulate specific user-defined objects.

AI/ML arXiv cs.AI

Modeling Day-Long ECG Signals to Predict Heart Failure Risk with Explainable AI

Develops DeepHHF, a deep learning model using 24-hour ECG data to predict heart failure risk with a high degree of accuracy and explainability.

AI/ML arXiv cs.AI

AI-enhanced tuning of quantum dot Hamiltonians toward Majorana modes

Proposes a neural network model to autotune quantum dot simulators toward Majorana modes using unsupervised learning on conductance maps.

AI/ML arXiv cs.AI

Movement Primitives in Robotics: A Comprehensive Survey

A comprehensive survey of movement primitives in robotics, detailing frameworks for representing and generating motor commands through human demonstrations.

AI/ML arXiv cs.AI

PiDR: Physics-Informed Inertial Dead Reckoning for Autonomous Platforms

Presents PiDR, a physics-informed inertial dead-reckoning framework that improves navigation accuracy for autonomous platforms without GNSS or visual data.

AI/ML arXiv cs.AI

Policy-Embedded Graph Expansion: Networked HIV Testing with Diffusion-Driven Network Samples

Introduces PEGE and DDB, a framework and diffusion-based model for improving the efficiency of network-based HIV testing in data-limited settings.

AI/ML arXiv cs.AI

Bi-Anchor Interpolation Solver for Accelerating Generative Modeling

Introduces the BA-solver, which uses a lightweight SideNet to accelerate generative Flow Matching models by reducing required neural function evaluations.

AI/ML arXiv cs.AI

Stabilizing the Q-Gradient Field for Policy Smoothness in Actor-Critic Methods

Proposes PAVE, a regularization framework for actor-critic methods that stabilizes the Q-gradient field to prevent erratic policy oscillations in physical robots.

Cybersecurity Hacker News

Aikido Code Audit

A discussion regarding a code audit performed by Aikido.

Other Hacker News

A record 242 US cities now have starter homes that cost $1M

Reports on the rising cost of starter homes in various US cities, noting that many now exceed $1M.

AI/ML arXiv cs.AI

Oranits: Mission Assignment and Task Offloading in Open RAN-based ITS using Metaheuristic and Deep Reinforcement Learning

Introduces Oranits, a system for mission assignment and task offloading in Open RAN-based intelligent transportation systems using metaheuristics and DRL.

AI/ML arXiv cs.AI

Charting the Future of Scholarly Knowledge with AI: A Community Perspective

Discusses the need for AI-enabled approaches and cross-disciplinary dialogue to better structure and synthesize scholarly knowledge.

AI/ML arXiv cs.AI

From Construction to Injection: Edit-Based Fingerprints for Large Language Models

Proposes an end-to-end fingerprinting framework using code-mixing and multi-candidate editing to protect LLMs from unauthorized redistribution.

AI/ML arXiv cs.AI

Enhancing Generative Auto-bidding with Offline Reward Evaluation and Policy Search

Presents AIGB-Pearl, a method integrating generative planning and policy optimization to enhance auto-bidding in advertising systems.

AI/ML arXiv cs.AI

RoboSSM: Scalable In-context Imitation Learning via State-Space Models

Introduces RoboSSM, a scalable in-context imitation learning framework for robots using state-space models (Longhorn) instead of Transformers.

AI/ML arXiv cs.AI

MENTOR: Reinforcement Learning via Flexible Teacher-Optimized Rewards for Tool-Use Distillation

Proposes MENTOR, a RL-based approach to distill tool-use capabilities from LLMs into smaller models using flexible teacher-optimized rewards.

AI/ML arXiv cs.AI

Beyond Reasoning Gains: Mitigating General-Capability Forgetting in Large Reasoning Models

Introduces RECAP, a replay strategy with dynamic objective reweighting to prevent capability forgetting in large reasoning models during post-training.

AI/ML arXiv cs.AI

Bid Farewell to Seesaw: Towards Accurate Long-tail Session-based Recommendation via Dual Constraints of Hybrid Intents

Proposes the HID framework to improve long-tail session-based recommendations by reducing the noise associated with low-exposure items.

Software Engineering Hacker News

Hey, N00B, We Didn't Hire You to Complete Tasks

A discussion on the expectations for junior developers, arguing that their role is more than just completing assigned tasks.

Software Engineering Hacker News

Digital Printing of Arabic: explaining the problem

An explanation of the technical challenges involved in the digital printing of Arabic text.