argumentA line of reasoning from premises to a conclusion. Like a claim, it belongs to no one who makes it.constitution →

The argument from settings where the student does not beat the supervisor

againstThis argument, if it holds, weighs against the claim.constitution →Strong pretrained language models fine-tuned on weak-model labels consistently outperform their weak supervisors.

The argument

The student's margin over its supervisor shrinks to nothing, or reverses, in identifiable settings. In the original experiments it was close to zero on chess with the smallest supervisors and recovered only about a tenth of the available gap on reward modeling. On hard generative reasoning, a Llama2-70b student fine-tuned on Gemma-2b's solutions to MATH problems scored below Gemma-2b, and GPT-3.5-Turbo fine-tuned on weak labels for two mathematics benchmarks is reported not to generalize. It is also asserted that when the supervisor was trained on a different distribution, students often finish below it. A regularity that fails under conditions weak supervision would ordinarily meet, namely harder tasks, larger capability gaps and unfamiliar data, does not hold consistently.

Does it work?

Does it follow?: slight even if the premises holdEven if the premises are true, they make the conclusion only a little more likely.constitution →
Do the premises hold?: 1 not separately assessed

Granting its premises, the argument shows that the margin can vanish and occasionally reverse, which tells against a reading of "consistently" as "always" but leaves the regularity standing if the exceptions are few. As the evidence stands they are few. The chess and reward-modeling results are small positive margins, not reversals. The MATH reversal is one pairing with a small difference, and the GPT-3.5-Turbo failure is reported without readable numbers. The premise that would carry most weight, that students often fall below supervisors trained on different data, is doubtful for language models: the main tables of the paper asserting it show the student ahead in every language setting, with one reversal on an image dataset.

Premises

UnassessedNo current assessment. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.constitution →When the weak supervisor was trained on a different data distribution, strong models naively fine-tuned on its labels often perform worse than the supervisor.

Where it is made

Passages where sources make, criticize or mention this argument. They record where it appears, not how strong it is, and none has precedence over another.

No source is recorded as making it.

Earlier ratings

Nov 11, 2026slight even if the premises holdEven if the premises are true, they make the conclusion only a little more likely.constitution →
Granting its premises, the argument shows that the margin can vanish and occasionally reverse, which tells against a reading of "consistently" as "always" but leaves the regularity standing if the exceptions are few. As the evidence stands they are few. The chess and reward-modeling results are small positive margins, not reversals. The MATH reversal is one pairing with a small difference, and the GPT-3.5-Turbo failure is reported without readable numbers. The premise that would carry most weight, that students often fall below supervisors trained on different data, is doubtful for language models: the tables of the paper asserting it show the student ahead in every language setting, with one reversal on an image dataset.
Nov 11, 2026slight even if the premises holdEven if the premises are true, they make the conclusion only a little more likely.constitution →
Granting its premises, the argument shows that the margin can vanish and occasionally reverse, which tells against a reading of "consistently" as "always" but leaves the regularity standing if the exceptions are few. As the evidence stands they are few. The chess and reward-modeling results are small positive margins, not reversals. The MATH reversal is one pairing with a difference inside sampling error, and the GPT-3.5-Turbo failure is reported without readable numbers. The premise that would carry most weight, that students often fall below supervisors trained on different data, is doubtful for language models: the tables of the paper asserting it show the student ahead in every language setting, with one reversal on an image dataset.