Pith. sign in

REVIEW 16 cited by

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

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2102.09690 v2 pith:MIUIVO5A submitted 2021-02-19 cs.CL cs.LG

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

classification cs.CL cs.LG
keywords prompttrainingexampleslanguagenearaccuracyacrossanswers
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original 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 pre-training 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.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 16 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Chain-of-Thought Prompting Elicits Reasoning in Large Language Models

    cs.CL 2022-01 accept novelty 9.0

    Chain-of-thought prompting, by including intermediate reasoning steps in few-shot examples, elicits strong reasoning abilities in large language models on arithmetic, commonsense, and symbolic tasks.

  2. Discovering Latent Knowledge in Language Models Without Supervision

    cs.CL 2022-12 conditional novelty 8.0

    An unsupervised technique extracts latent yes-no knowledge from language model activations by locating a direction that satisfies logical consistency properties, outperforming zero-shot accuracy by 4% on average acros...

  3. UA-Legal-Bench: A Benchmark for Evaluating Large Language Models on Ukrainian Legal Reasoning

    cs.CL 2026-05 unverdicted novelty 7.0

    UA-Legal-Bench is a new five-task benchmark for Ukrainian legal reasoning that demonstrates task-dependent few-shot prompting effects and the need for macro-F1 over accuracy on imbalanced classes.

  4. PRIME: Evaluating Prompt Resolution Under Incompatible Instructions in LLMs

    cs.AI 2026-06 unverdicted novelty 6.0

    PRIME is a new evaluation framework that creates calibrated conflicts in LLM prompts and finds conflict type affects model behavior more than scale.

  5. The Tokenizer Tax Across 25 European Languages: Domain Invariance, Cross-Lingual Few-Shot Effects, and the Ukrainian Penalty

    cs.CL 2026-05 unverdicted novelty 6.0

    Tokenizer fertility varies 2.5x across 25 European languages with domain-invariant rankings, morphological fragmentation in high-fertility cases, and a Ukrainian penalty from pre-training underrepresentation.

  6. AMEL: Accumulated Message Effects on LLM Judgments

    cs.AI 2026-05 unverdicted novelty 6.0

    LLMs exhibit an accumulated message effect where conversation history polarity biases subsequent judgments, stronger for high-entropy items, independent of context length, and with a negativity bias.

  7. AMEL: Accumulated Message Effects on LLM Judgments

    cs.AI 2026-05 conditional novelty 6.0

    LLMs exhibit an accumulated message effect where conversation history saturated with positive or negative evaluations biases subsequent judgments, with larger shifts on uncertain items, a negativity asymmetry, and no ...

  8. Leveraging LLMs for Multi-File DSL Code Generation: An Industrial Case Study

    cs.SE 2026-04 unverdicted novelty 6.0

    Fine-tuning 7B code LLMs on a custom multi-file DSL dataset achieves structural fidelity of 1.00, high exact-match accuracy, and practical utility validated by expert survey and execution checks.

  9. Explicit Logic Channel for Validation and Enhancement of MLLMs on Zero-Shot Tasks

    cs.AI 2026-03 conditional novelty 6.0

    An Explicit Logic Channel of LLM, VFM and probabilistic inference validates and improves zero-shot MLLMs via Consistency Rate without ground-truth labels.

  10. Simple synthetic data reduces sycophancy in large language models

    cs.CL 2023-08 unverdicted novelty 6.0

    Scaling and instruction tuning increase sycophancy in LLMs on opinion and fact tasks, but a synthetic data fine-tuning intervention reduces it on held-out prompts.

  11. Emergent Abilities of Large Language Models

    cs.CL 2022-06 unverdicted novelty 6.0

    Emergent abilities are capabilities present in large language models but absent in smaller ones and cannot be predicted by extrapolating smaller model performance.

  12. When Context Sticks: Studying Interference in In-Context Learning

    cs.LG 2026-04 unverdicted novelty 5.0

    In-context learning shows persistent interference from prior examples, with more misleading linear examples degrading quadratic predictions and training curricula modulating recovery speed.

  13. Consistency Analysis of Sentiment Predictions using Syntactic & Semantic Context Assessment Summarization (SSAS)

    cs.CL 2026-04 unverdicted novelty 5.0

    SSAS improves LLM sentiment prediction consistency and data quality by up to 30% on three review datasets via syntactic and semantic context assessment summarization.

  14. Explicit Logic Channel for Validation and Enhancement of MLLMs on Zero-Shot Tasks

    cs.AI 2026-03 unverdicted novelty 5.0

    Introduces Explicit Logic Channel (ELC) with LLM, VFM and probabilistic inference for validating, selecting and enhancing MLLMs on zero-shot tasks using Consistency Rate and cross-channel integration.

  15. (Towards) Scalable Reliable Automated Evaluation with Large Language Models

    cs.CL 2026-07 conditional novelty 4.0

    Multi-LLM pairwise Elo ranking with adjustable consensus thresholds produces rankings of competency profiles that average Spearman ρ≈0.83 with expert judgments.

  16. Supervision versus Demonstration-Based In-Context Learning for Multiword Expression Classification

    cs.CL 2026-06 unverdicted novelty 4.0

    On a controlled Turkish dataset of 147 examples, few-shot prompting lets some LLMs match or beat a supervised BERT baseline for LVC detection, though results are highly sensitive to prompt design.