All Articles
15956 articles total
Agreement Is Not Quality: Blind Expert Verification of Human and LLM Qualitative Coding When Human Consensus Is Not Ground Truth
This research challenges the use of human agreement as the gold standard for LLM qualitative coding, showing that blind expert verification can favor LLM interpretations over human consensus.
TORUS: A Test of Rendering-Understanding Self-Coherence for Unified Audio Models
TORUS is introduced as a self-coherence test for unified audio models to determine if their audio generation and understanding capabilities are aligned.
Design Concept: Scaffolding Geopolitical Reflection Among Tech Workers
A speculative HCI design proposal for an AI-enabled narrative system aimed at encouraging geopolitical reflection among technology workers.
Gated Q-learning: Add Off-Policy Bias to Taste
Gated Q-learning is proposed as a framework to manage off-policy bias in reinforcement learning by interpolating between eliminating bias and ignoring it via a gating mechanism.
FairFund-Bench: Evaluating Distributive Bias in LLM Resource Allocation
FairFund-Bench is a new benchmark for evaluating distributive bias in LLM resource allocation, revealing that audit formats significantly influence bias detection.
DiffAttack: Evasion Attacks Against Face Recognition via Latent Diffusion Models
DiffAttack uses latent diffusion models to create high-quality adversarial images that can evade facial recognition systems with high success rates.
Retrieval-Driven Training-Free AI-Generated Video Attribution
A training-free attribution paradigm is introduced to identify the generative source of AI-generated videos using a generative fingerprint-based pipeline.
Less Coffee, Better Sleep
A discussion on lifestyle choices regarding caffeine consumption and its impact on sleep quality.
What DMARC Protects You From, and What It Does Not
An explanation of the DMARC protocol and its specific capabilities and limitations in email security.
MPs demand answers on Fujitsu's inclusion in lucrative frameworks
Political scrutiny in the UK regarding Fujitsu's inclusion in government frameworks.
PISIGuard: Protect your personal and sensitive info when you chat with AI
A project called PISIGuard designed to protect sensitive information during AI interactions.
The true power of regular expressions (2012)
A technical retrospective on the power and utility of regular expressions.
Show HN: We Fixed UniFi's Slow PPPoE Performance with PPPoE Half-Bridge
A developer-focused fix for slow UniFi PPPoE performance using a half-bridge method.
A Marc Benioff-backed startup thinks AI can solve the AI deployment problem
A startup named June emerged from stealth with significant funding to simplify AI deployment.
A Unified Benchmark of Deep Learning Models for Multi-task 3D Brain Tumor Segmentation from Magnetic Resonance Imaging
A research paper providing a unified benchmark for deep learning models used in 3D brain tumor segmentation.
TextCloak: Thwarting Unauthorized LLM Exploitation via RL-Driven Unlearnable Text
A research paper proposing TextCloak, an RL-driven framework to prevent unauthorized LLM exploitation using unlearnable text.
Validation Evidence in LLM Repair Agents: How Much of What Passes Actually Tests the Bug?
A study analyzing the effectiveness of validation evidence in LLM-driven software repair agents.
Octane – React's programming model, compiled
An exploration of Octane, a project that compiles React's programming model for potentially better performance.
WitCert: Sound Runtime Risk Observability and Gating for KV-Cache Quantization
Introduces WitCert, a runtime observability tool for KV-cache quantization in LLMs that provides provable bounds on compression damage.
A user's guide to PINNs in geometric analysis: lessons from the asymptotic Plateau problem
A guide on using Physics-Informed Neural Networks (PINNs) for geometric analysis, detailing optimizations that reduce training costs by 40-50x.