Skip to content
securesein
Model UpdatesResearchBenchmarksSecurityFundamentalsTopicsAbout

Topics

Seven shelves, deliberately few. Each one pulls together everything on that subject — the two-minute news item and the twenty-minute teardown sit side by side.

  • LLMs

    Language-model architecture, behaviour and capability.

    23 postsLatest 14 Sept 2026

    Understanding Adversarial Attacks on Large Language Models

  • Reasoning

    Chain of thought, test-time compute, reasoning models and their limits.

    8 postsLatest 13 Sept 2026

    Rethinking Communication: When Reasoning Meets Information Theory

  • Agents

    Tool use, agent frameworks, MCP, multi-step autonomy.

    14 postsLatest 19 Sept 2026

    Proving Instead of Testing

  • Interpretability

    Features, circuits, probing, activation steering, representation engineering.

    1 postLatest 5 Sept 2026

    Abliteration: how one direction holds a model's refusals

  • Deep learning

    Architectures, optimisation, scaling laws, training dynamics.

    5 postsLatest 17 Sept 2026

    Challenging AI Claims: A New Protocol for Transparent and Falsifiable Records

  • Reinforcement learning

    RLHF, RLVR, reward models, policy optimisation.

    1 postLatest 12 Sept 2026

    Curiosity at the Edge of Chaos: What a New RL Paper Actually Shows

  • Multimodal

    Vision, audio, video and cross-modal models.

    4 postsLatest 17 Sept 2026

    Bridge Over Troubled Layers: Multi-Agent Detection of Logic Vulnerabilities in Blockchain-IoT

  • Training

    Pre-training, fine-tuning, data, tokenisation, curricula.

    4 postsLatest 17 Sept 2026

    Unmasking the Automaton: How Additive Input Pathways Skew State Tracking in Linear RNNs

  • Inference

    Serving, quantisation, speculative decoding, caching, local and on-device.

    4 postsLatest 9 Sept 2026

    Frontier Reasoning on the Edge: Deploying and Optimizing Models with NVIDIA Jetson

  • Evaluation

    Benchmark design, contamination, harnesses, construct validity.

    2 postsLatest 12 Sept 2026

    How Do You Catch a Machine in a Lie?

  • AI security

    Prompt injection, jailbreaks, agent hijacking, exfiltration, model supply chain.

    11 postsLatest 17 Sept 2026

    Bridge Over Troubled Layers: Multi-Agent Detection of Logic Vulnerabilities in Blockchain-IoT

  • AI safety

    Alignment, guardrails, deception, misuse, model behaviour under pressure.

    6 postsLatest 19 Sept 2026

    Proving Instead of Testing

  • Generative AI

    Diffusion, image, audio and video generation, and their systems.

    3 postsLatest 17 Sept 2026

    Fortifying Industrial Safety AI with Synthetic Data on Amazon SageMaker

securesein

securesein

Keeping up with machine learning, one mechanism at a time.

Read

  • Model Updates
  • Research
  • Benchmarks
  • Security
  • Fundamentals
  • Topics
  • Threads

Feeds

  • Everything
  • Model Updates
  • Research
  • Benchmarks
  • Security
  • Fundamentals

This site

  • About
  • Colophon
  • Search

© 2026 securesein. Curated by a human, drafted in part by an AI —which part, and by which model.