Cover image

A Picture Is Not a Permission Boundary

TL;DR for operators An AI agent reading an external document may encounter instructions that the document’s author has no authority to issue. Pictionary offers an unusual defense: convert that untrusted text into an image before the model reads it. The content remains available, but several tested models become substantially less likely to obey embedded commands. ...

November 2, 2026 · 8 min · Zelina
Cover image

Relevant History, Missing Rules: The Retrieval Risk in AI Assistants

TL;DR for operators An AI assistant can retrieve conversation history more efficiently and become less reliable at following instructions. The critical distinction is between standing rules that remain binding and past exchanges whose relevance changes with each question. Role-aware Heuristic Episodic Attention (REA) addresses both requirements through separate instruction and episodic memories. In its experiments, the complete architecture improves long-conversation quality while substantially reducing response latency. Yet removing the dedicated instruction memory causes instruction adherence to deteriorate sharply when similarity-based retrieval is introduced. ...

October 31, 2026 · 9 min · Zelina
Cover image

Following Instructions Is Not the Same as Knowing More

TL;DR for operators A multimodal model can become much better at obeying instructions without becoming much better at the underlying tasks those instructions govern. In the experiments examined here, one 8B vision-language model gains 10.58 percentage points on a targeted instruction-following benchmark and another gains 22.91 points. Yet their average results across broader STEM, VQA, OCR, and document-understanding tests move by only +0.33 and -0.21 points. ...

October 1, 2026 · 7 min · Zelina
Cover image

Let the Model Design the Poster—Not the Evidence

TL;DR for operators A scientific poster can look polished and still fabricate the plots or diagrams readers interpret as evidence. PosterHarness separates those responsibilities: the image model designs the layout and decides where evidence should appear, but it must leave those regions blank for source-paper figures to be inserted later by deterministic code. ...

August 3, 2026 · 8 min · Zelina
Cover image

Skill Issue or System Design? How LLMs Actually Follow Instructions

The checklist problem that exposes the model Checklist tasks look boring. That is exactly why they are useful. Ask an LLM to write a formal email under 50 words, include one required term, avoid another term, and return the result as JSON. None of this sounds intellectually difficult. No theorem proving. No multimodal reasoning. No dramatic benchmark leaderboard screenshot. Just instructions. ...

April 8, 2026 · 18 min · Zelina
Cover image

Replay the Losses, Win the Game: When Failed Instructions Become Your Best Training Data

Failure logs are usually treated as evidence for the prosecution. A model is asked to produce a concise compliance summary with three bullet points, mention two risks, avoid prohibited claims, and end with a recommendation. It produces three bullets, correctly identifies the risks, avoids the prohibited claims—and forgets the recommendation. Under a strict binary reward, the response receives a zero. Under a partial-credit reward, it might receive 0.75. The first signal says nothing useful happened. The second says something useful happened, but not precisely what. ...

December 30, 2025 · 18 min · Zelina