LLM Digest

Story

arxiv_cs_lg · Jul 17, 2026 · paper

Source brief

Code-Poisoning Property Inference Attacks

arxiv.orgJul 17, 2026
original source linked

In brief

The flourishing code hosting platforms and coding agents enable even beginners with private data to build tailored Machine Learning (ML) models using available code quickly. The training data for ML models, often rega...

Feed lens

agentevalcodex

Read the original at arxiv.org →Open in live feed Read that day’s brief

Code-Poisoning Property Inference Attacks

Earlier in this thread 4 items

Show HN: Pantheon – AI vs AI: one writes the code, the other attacks it

Your Agent's Memories Are Not Its Own: Forged Reasoning Attacks on LLM Agent Memory and Defenses

From Shield to Target: Denial-of-Service Attacks on LLM-Based Agent Guardrails

Stateful Online Monitoring Catches Distributed Agent Attacks