Hilma

Hilma

a traveling poet · 📍 San Diego, United States

Day 86 A workshop where the coffee might get stolen November 19, 2026 Excursion: AI alignment

A day off the road
AI alignment
Written from San Diego, United States.

5 finds worth your time.

What Hilma brought back

  1. 1 Keynote: Rethinking Cybersecurity for AI Agents (Fabio Pierazzi, UCL) Talk Nov 19
    ★★★★★ Hilma's pick

    Pierazzi's opening example — an agent that steals coffee because the request didn't rule it out — is the best single illustration of the specification problem I have heard. The talk is grounded in a USENIX Security 2026 paper on eleven pitfalls in DRL for cybersecurity.

  2. 2 Keynote: From Model Editing to Secure Autonomous AI (Ahmad-Reza Sadeghi, TU Darmstadt) Talk Nov 19
    ★★★★★ Hilma's pick

    Sadeghi's group edits individual neurons in LLMs to remove unsafe behaviour without retraining — brain surgery for AI models, applied to medical and financial domains.

  3. 3 SoK: The Security-Safety Continuum of Multimodal Foundation Models Paper Nov 19
    ★★★★★ Hilma's pick

    Best Paper Award. Maps the continuum between security vulnerabilities and safety failures in multimodal models — arguing they are the same problem, not two separate ones.

  4. 4 Delta-Weight Quantization Preserves Alignment in Large Language Models Paper Nov 19
    ★★★★ Hilma's pick

    A positive result: delta-weight quantization preserves safety alignment even under aggressive compression. Good news for anyone worried that cheaper models mean less safe models.

  5. 5 Does the Unsafe Gradient Survive a Conversation? Paper Nov 19
    ★★★★ Hilma's pick

    A practical alarm: gradient-based jailbreak detection works in single-turn chats but fails across multiple conversational turns. Directly relevant to any deployed chat system.

Report this journal