All Articles
17090 articles total
Adaptive Model Compression (AMC): Saliency-Driven Resource Allocation for Ultra-Low-Power Transformer Inference
The Adaptive Model Compression (AMC) framework reduces energy consumption by 59.2% and increases throughput by 2.24x for transformer inference on edge devices.
Minionese: Comprehensive Benchmark and Mechanistic Study of Multilingual LLM Safety
Introduction of Minionese, a benchmark for evaluating multilingual LLM safety and jailbreak vulnerabilities across 18 languages.
Launch HN: Agnost AI (YC S26) – Extract user feedback from agent conversations
Agnost AI is a new YC-backed tool designed to extract user feedback from conversations with AI agents.
Superoptimizer – A Look at the Smallest Program
An exploration of a 'superoptimizer', a program designed to find the smallest possible version of a given program.
The Agentic Loop: Three loops in a trench coat
A discussion analyzing the conceptual structure of 'agentic loops' in AI systems.
Meta’s Adam Mosseri says AI token budgets could soon be capped per engineer
Meta's Adam Mosseri suggests that companies may implement token budgets for engineers to control AI spending.
The OLED Xbox Ally X20 is so good, Asus will sell it solo
Asus will release a standalone version of the OLED Xbox Ally X handheld, moving away from a mandatory AR glasses bundle.
New York bans data center construction for a year, rattling AI industry
New York has imposed a one-year moratorium on new data center construction, impacting the AI industry's infrastructure growth.
Robust, Scalable Detection of Text Containment in Large Web-Crawled Corpora
Introduction of FindMyText, an open-source Python package for detecting near-verbatim text containment in large web-crawled corpora.
Geometric mean-based pairwise comparison method with the reference values -- statistical approach
A statistical approach to the pairwise comparison method using geometric means to improve decision-making weight calculations.
Quantum Circuit Vision: Cost-Aware Evaluation of Visual AI Agents for Quantum Code Generation
Presentation of Quantum Circuit Vision, a cost-aware evaluation framework for AI agents generating quantum code from circuit diagrams.
I'm a USB-C Maximalist
A Hacker News discussion thread where the user discusses their preference for using USB-C as a universal standard for all connectivity.
Google Images gets a Pinterest-like redesign focused on discovery
Google Images is introducing a Pinterest-style redesign to encourage discovery through a personalized 'For You' gallery.
The Google Images homepage will recommend photos even before you search
Google is updating its Images homepage to show recommended photos based on user interests before a search is even performed.
Spotify is now an AI chatbot, too
Spotify is testing a new AI chatbot called 'Talk to Spotify' that allows Premium users to explore music and podcasts via conversation.
A Foundation Model for Multimodal Event Sequences in Financial Applications
Researchers present a foundation transformer model for multimodal event sequences in financial applications to replace task-specific models.
Workload-Driven Optimization for On-Device Real-Time Subtitle Translation
A study on optimizing on-device English-to-Chinese subtitle translation by reducing vocabulary size and utilizing GGUF quantization for faster inference.
Learning in Curved Weight Space:Exponential-Linear Weight Reparameterization for Improved Optimization
Introduction of Exponential-Linear Weight Reparameterization (ELWR) to improve neural network optimization speed by combining linear and exponential pathways.
A Production-Oriented Framework for Evaluation of SFX Generation
A new production-oriented framework for evaluating sound effect (SFX) generation, comparing methods like AudioX for reference alignment and diversity.
An LLM-powered Agentic Recommendation System for Connected TV Content Discovery
A hybrid agentic recommendation system for Connected TV that combines LLM reasoning with traditional ML for efficient content discovery.