theNotebook

The Notebook · the week

The week of 5 October 2026

7 ideas · 5 October 2026 to 11 October 2026

  1. Monday · 5 October 2026

    nobody asks the whole question on the first turn

    A study published this week took multi-hop questions from the usual benchmarks and broke each one into the sort of conversation people actually have – a vague opening, a follow-up, then the real question somewhere near …

    For a ten-year-oldFor an architect

  2. Tuesday · 6 October 2026

    the rehearsal went well because nobody else was in it

    A benchmark out last week turned language-model agents loose on 119 chess endgames where a forced mate exists, and handed them a console with three separate verbs: make a real move, ask about the board, or simulate …

    For a ten-year-oldFor an architect

  3. Wednesday · 7 October 2026

    the backslash went missing somewhere between the model and the shell

    Someone read 47,828 shell commands out of real agent sessions this week and found that the harness, not the model, had quietly rewritten about one call in eight among those carrying code or escape characters, and that …

    For a ten-year-oldFor an architect

  4. Thursday · 8 October 2026

    the rule at the top of the query turned out to be more query

    Eight embedding models, a plain instruction at the top of each query, and one small change to the test: put a paraphrase of the query in alongside the query itself, something nobody asked to retrieve.

    For a ten-year-oldFor an architect

  5. Friday · 9 October 2026

    the steadiest judge was wrong the same way every time

    A paper out this week audited six frontier models doing the job most of us quietly hand them, scoring other models' answers, and the numbers are worth reading twice.

    For a ten-year-oldFor an architect

  6. Saturday · 10 October 2026

    the thinking-off switch turned out to be a suggestion

    Somebody finally went and looked at the other half of the thinking switch. A paper this week takes six models and six different ways of saying don't reason, just answer, and then declines to trust any of the flags.

    For a ten-year-oldFor an architect

  7. Sunday · 11 October 2026

    the limits were in the file the agent was allowed to edit

    A paper out this week asks a question most agent runtimes answer by accident: which parts of its own configuration may an agent rewrite? Some fields grant abilities, like switching a tool on.

    For a ten-year-oldFor an architect

Search the notebook