Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/colm-writing-style
Use when drafting or revising COLM paper prose — leading with a finding about language models rather than a leaderboard delta, scoping claims to tested models and scales, naming versions in text, keeping the 9-page main text self-sufficient, and matching the measured, analysis-forward voice of COLM's award lineage.
Passed 21 of 22 test cases with the skill, and 10 of 22 without it.
and 18 more in the full report
12 fixed9 passed either way0 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 12 test cases the skill fixed. Same model, same prompt, with and without the skill.
Could you critique this paper section comparing instruction-tuned models against base checkpoints before I submit to COLM? Please analyze the draft and structure your feedback using these specific breakdown fields: Register (identifying measurement tone vs marketing residue), Finding sentence (the core claim about model behavior), Scope violations (tally and worst case instance), Unpinned mentions (models missing concrete version strings), 9-page audit (flagging core claims reliant on appendices), and Edit order (list of passes left to run).
| Checked | Without skill | With skill |
|---|---|---|
| The response includes the bracketed field key [Register]. | Missed | Passed |
| The response includes the bracketed field key [Scope violations]. | Missed | Passed |
| The response includes the bracketed field key [9-page audit]. | Missed | Passed |
The instructions your model reads when the skill loads.
COLM prose has a recognizable register, visible in the venue's Outstanding Papers: it reads like measurement, not marketing. The model under study is named precisely, the claim is scoped to what was tested, and the interesting sentence is about what was learned, not who is winning. Edit toward that register.
The weakest COLM opening is "LLMs have shown impressive capabilities, but..." — a frame that positions the paper as a reaction to hype. The strongest openings state the phenomenon or question directly:
If the paper's most interesting sentence is a number on a benchmark, the framing is not done; find the sentence about language models that the number supports.
Unscoped generalization is the most common COLM prose bug. Claims about "LLMs" must either carry coverage (multiple families, multiple scales) or shrink to their evidence:
| Draft sentence | Scoped rewrite |
|---|---|
| "LLMs cannot do multi-step planning." | "None of the six models tested (App. A) exceeds 40% on depth-3 planning items." |
| "Instruction tuning destroys calibration." | "Across both model families, instruction-tuned variants are less calibrated than their bases at every scale we test." |
| "GPT-4 solves this task." | " (queried 2026-03-1x) solves 87% of instances under the App. B prompt." |
| "Our method generalizes to any decoder." | "The method assumes access to token logprobs; it applies to open-weight decoders and APIs that expose them." |
Version strings and query dates belong in the text or a footnote at first mention, not only in an appendix — at this venue they are part of the noun.
test cases passed, vs 10 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.