SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2102.09690 · ICML · 2021

Calibrate Before Use: Improving Few-Shot Performance of Language Models

Dan Klein, Eric Wallace, Tony Zhao, Shi Feng, Sameer Singh

arXiv · PDF · Open in the Atlas

Code that ran

None of the 4 samples (functions and classes) lifted from this paper's linked repositories has a recorded run on a synthesized input (4 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2102.09690.json · badge: /api/ran/2102.09690.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
tonyzhaozh/few-shot-learning canonical (classifier label)
classifier basis: classifier_2025
0 ran · 4 no recorded run
FunctionStatusWhere it lives
em_accuracy_helper Not yet run
unverified (harvested, no recorded run)
tonyzhaozh/few-shot-learning/run_extraction.py
code served (permissive licence) · get_code("e1d89d7579eaaeec")
How it ran

no invocation recorded.

eval_accuracy Not yet run
unverified (harvested, no recorded run)
tonyzhaozh/few-shot-learning/run_classification.py
code served (permissive licence) · get_code("7753000f9482cac5")
How it ran

no invocation recorded.

get_label_probs Not yet run
unverified (harvested, no recorded run)
tonyzhaozh/few-shot-learning/run_classification.py
code served (permissive licence) · get_code("77347b697708c203")
How it ran

no invocation recorded.

get_p_content_free Not yet run
unverified (harvested, no recorded run)
tonyzhaozh/few-shot-learning/run_classification.py
code served (permissive licence) · get_code("d09226e56c2085ea")
How it ran

no invocation recorded.

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 0 the same month, 2 later) and 0 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
get_label_probs 3 other papers
1 dated earlier, 2 later
2005.14165 (2020-05)
Language Models are Few-Shot Learners
2401.12406 (2024-01)
Enhancing In-context Learning via Linear Probe Calibration
2411.02988 (2024-11)
Confidence Calibration of Classifiers with Many Classes
get_p_content_free 2 other papers
1 dated earlier, 1 later
2005.14165 (2020-05)
Language Models are Few-Shot Learners
2312.04021 (2023-12)
A Study on the Calibration of In-context Learning
em_accuracy_helper 1 other paper
1 dated earlier
2005.14165 (2020-05)
Language Models are Few-Shot Learners
eval_accuracy 1 other paper
1 dated earlier
2005.14165 (2020-05)
Language Models are Few-Shot Learners

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

GPT-3 can perform numerous tasks when provided a natural language prompt that contains a few training examples. We show that this type of few-shot learning can be unstable: the choice of prompt format, training examples, and even the order of the training examples can cause accuracy to vary from near chance to near state-of-the-art. We demonstrate that this instability arises from the bias of language models towards predicting certain answers, e.g., those that are placed near the end of the prompt or are common in the pretraining data. To mitigate this, we first estimate the model's bias towards each answer by asking for its prediction when given the training prompt and a content-free test input such as "N/A". We then fit calibration parameters that cause the prediction for this input to be uniform across answers. On a diverse set of tasks, this contextual calibration procedure substantially improves GPT-3 and GPT-2's average accuracy (up to 30.0% absolute) and reduces variance across different choices of the prompt.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2102.09690")
get_code_for_paper("2102.09690")
have("2102.09690")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2102.09690.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2102.09690.svg)](https://syntology.ai/paper/2102.09690)

Connect an agent — have() is free.