arXiv News

First-hand information for everyone

Site

AboutPrivacy PolicyTerms of Use

© 2026 arXiv News

arXiv News

EnglishJapanese

Switch language

EnglishJapanese
Loading account…

First-hand information for everyone

Latest
Pure differential privacy for many queries at the square‑root error rate — a long‑standing conjecture provedAuthentication in quantum networks: a clear survey of what’s needed and why setup mattersA safety 'kernel' that the AI cannot turn off enforces actions at runtimeA concrete quantum circuit for “one‑shot” digital signaturesAVISE: an open framework that automates finding jailbreaks in language modelsAVISE: an open‑source framework that automates tests for AI security and jailbreaksQuantum entropy to tiny devices: a service that sends QRNG randomness to ESP32s over post-quantum linksGenomic language models can memorize DNA — new framework measures privacy riskFLA3: A federated learning system that enforces who, when and why across international healthcare studiesNew study maps 193 security threats for multi‑agent AI and finds major gaps in existing frameworksPure differential privacy for many queries at the square‑root error rate — a long‑standing conjecture provedAuthentication in quantum networks: a clear survey of what’s needed and why setup mattersA safety 'kernel' that the AI cannot turn off enforces actions at runtimeA concrete quantum circuit for “one‑shot” digital signaturesAVISE: an open framework that automates finding jailbreaks in language modelsAVISE: an open‑source framework that automates tests for AI security and jailbreaksQuantum entropy to tiny devices: a service that sends QRNG randomness to ESP32s over post-quantum linksGenomic language models can memorize DNA — new framework measures privacy riskFLA3: A federated learning system that enforces who, when and why across international healthcare studiesNew study maps 193 security threats for multi‑agent AI and finds major gaps in existing frameworks

Today's Briefing

Friday, July 31, 2026
AllArtificial IntelligenceMachine LearningNatural Language ProcessingComputer VisionRoboticsCryptographyPhysicsMathematics
CryptographyFeatured briefing

Pure differential privacy for many queries at the square‑root error rate — a long‑standing conjecture proved

Researchers have resolved a question posed by Nikolov and Ullman about how accurately one can answer many statistical queries while satisfyi

July 23, 2026EN2 min read
Read full article

Latest Research

Cryptography
June 30, 2026

Authentication in quantum networks: a clear survey of what’s needed and why setup matters

This paper reviews how to prove who sent data and that it was not changed in networks that carry quantum information. The authors separate t

EN
2 min read
Artificial Intelligence
June 25, 2026

A safety 'kernel' that the AI cannot turn off enforces actions at runtime

This paper describes the Unfireable Safety Kernel, a separate safety module that runs outside an AI agent so the agent cannot bypass it. The

EN
2 min read
Cryptography
June 23, 2026

A concrete quantum circuit for “one‑shot” digital signatures

Researchers describe a full quantum-circuit implementation of a one‑shot signature scheme. In this kind of scheme, a signing key is a fragil

EN
2 min read
Advertisement
Artificial Intelligence
April 24, 2026

AVISE: an open framework that automates finding jailbreaks in language models

Researchers introduce AVISE (AI Vulnerability Identification and Security Evaluation), a modular open-source framework to find security prob

EN
2 min read
Artificial Intelligence
April 23, 2026

AVISE: an open‑source framework that automates tests for AI security and jailbreaks

This paper introduces AVISE (AI Vulnerability Identification and Security Evaluation), a modular, open‑source framework meant to help resear

EN
2 min read
Cryptography
March 16, 2026

Quantum entropy to tiny devices: a service that sends QRNG randomness to ESP32s over post-quantum links

This paper describes a working system that moves high-quality random bytes produced by a quantum random number generator (QRNG) to small Int

EN
2 min read
Machine Learning
March 14, 2026

Genomic language models can memorize DNA — new framework measures privacy risk

Researchers tested how much genomic language models (GLMs) can memorize specific DNA sequences and how that creates privacy risks. They buil

EN
2 min read
Cryptography
March 13, 2026

FLA3: A federated learning system that enforces who, when and why across international healthcare studies

Researchers present FLA3, a federated learning platform designed to enforce governance rules during multi‑centre clinical research. The pape

EN
2 min read
Artificial Intelligence
March 13, 2026

New study maps 193 security threats for multi‑agent AI and finds major gaps in existing frameworks

This paper looks at security risks that arise when many autonomous AI agents work together. These “multi‑agent systems” share memory, call e

EN
2 min read
Next page of briefings