mlsanityData to 14 Oct 2026

Enhancing In-context Learning via Linear Probe Calibration

arXiv 2401.12406 ↗First posted 22 Jan 2024cs.CL cs.AI cs.LGcited by 16

On the map: Mechanistic interpretability → Activation steering methods

Abstract (arXiv)

In-context learning (ICL) is a new paradigm for natural language processing that utilizes Generative Pre-trained Transformer (GPT)-like models. This approach uses prompts that include in-context demonstrations to generate the corresponding output for a new query input. However, applying ICL in real cases does not scale with the number of samples, and lacks robustness to different prompt templates and demonstration permutations. In this paper, we first show that GPT-like models using ICL result in unreliable predictions based on a new metric based on Shannon entropy. Then, to solve this problem, we propose a new technique called the Linear Probe Calibration (LinC), a method that calibrates the model's output probabilities, resulting in reliable predictions and improved performance, while requiring only minimal additional samples (as few as five labeled data samples). LinC significantly enhances the ICL test performance of GPT models on various benchmark datasets, with an average improvement of up to 21%, and up to a 50% improvement in some cases, and significantly boosts the performance of PEFT methods, especially in the low resource regime. Moreover, LinC achieves lower expected calibration error, and is highly robust to varying label proportions, prompt templates, and demonstration permutations. Our code is available at urlhttps://github.com/mominabbass/LinC.

Key facts. Cited by 16 papers in the map (data through 14 Oct 2026). Placed in Activation steering methods, one of 13 directions in Mechanistic interpretability. Builds on 49 of its 67 references that are in the map. First posted on arXiv 22 Jan 2024.

Builds on

49 of 67 references.

References · 49

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  4. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  5. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  6. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  7. 2020Scaling Laws for Neural Language Modelscited by 6,586
  8. 2019BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehensioncited by 5,893
  9. 2017On Calibration of Modern Neural Networkscited by 4,960
  10. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  11. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,795
  12. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,780
37 more
  1. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,512
  2. 2019XLNet: Generalized Autoregressive Pretraining for Language Understandingcited by 2,811
  3. 2015Character-level Convolutional Networks for Text Classificationcited by 2,330
  4. 2019SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systemscited by 1,729
  5. 2021Multitask Prompted Training Enables Zero-Shot Task Generalizationcited by 1,470
  6. 2021Calibrate Before Use: Improving Few-Shot Performance of Language Modelscited by 1,392
  7. 2021Scaling Language Models: Methods, Analysis & Insights from Training Gophercited by 1,202
  8. 2021What Makes Good In-Context Examples for GPT-3?cited by 1,195
  9. 2021Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivitycited by 1,135
  10. 2021An Explanation of In-context Learning as Implicit Bayesian Inferencecited by 938
  11. 2017A Survey on Multi-Task Learningcited by 840
  12. 2022Out of One, Many: Using Language Models to Simulate Human Samplescited by 752
  13. 2022Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distributioncited by 739
  14. 2022Transformers learn in-context by gradient descentcited by 732
  15. 2022What Can Transformers Learn In-Context? A Case Study of Simple Function Classescited by 715
  16. 2022GPT-NeoX-20B: An Open-Source Autoregressive Language Modelcited by 713
  17. 2019Unified Language Model Pre-training for Natural Language Understanding and Generationcited by 670
  18. 2021Carbon Emissions and Large Neural Network Trainingcited by 567
  19. 2014OpenML: networked science in machine learningcited by 526
  20. 2019Quantifying the Carbon Emissions of Machine Learningcited by 521
  21. 2021MetaICL: Learning to Learn In Contextcited by 454
  22. 2022Data Distributional Properties Drive Emergent In-Context Learning in Transformerscited by 309
  23. 2022Prompting GPT-3 To Be Reliablecited by 260
  24. 2020On the Stability of Fine-tuning BERT: Misconceptions, Explanations, and Strong Baselinescited by 255
  25. 2023Transformers as Algorithms: Generalization and Stability in In-context Learningcited by 222
  26. 2019A Constructive Prediction of the Generalization Error Across Scalescited by 210
  27. 2021Meta-learning via Language Model In-context Tuningcited by 164
  28. 2022LIFT: Language-Interfaced Fine-Tuning for Non-Language Machine Learning Taskscited by 148
  29. 2022Diverse Demonstrations Improve In-context Compositional Generalizationcited by 139
  30. 2023Batch Calibration: Rethinking Calibration for In-Context Learning and Prompt Engineeringcited by 94
  31. 2022Prototypical Calibration for Few-shot Learning of Language Modelscited by 58
  32. 2022Sharp-MAML: Sharpness-Aware Model-Agnostic Meta Learningcited by 58
  33. 2023kNN Prompting: Beyond-Context Learning with Calibration-Free Nearest Neighbor Inferencecited by 47
  34. 2022Mitigating Gender Bias in Distilled Language Models via Counterfactual Role Reversalcited by 40
  35. 2023Generative Calibration for In-context Learningcited by 20
  36. 2022Understanding Benign Overfitting in Gradient-Based Meta Learningcited by 11
  37. 2023Meta-Learning with a Geometry-Adaptive Preconditionercited by 11

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 7; 2025: 8; 2026: 1.

16 papers.

Sort:

By area:

  1. 2024Thermometer: Towards Universal Calibration for Large Language Modelscited by 38
  2. 2025Restoring Calibration for Aligned Large Language Models: A Calibration-Aware Fine-Tuning Approachcited by 25
  3. 2024On Calibration of LLM-based Guard Models for Reliable Content Moderationcited by 23
  4. 2024Confidence Calibration of Classifiers with Many Classescited by 11
  5. 2025Solving Empirical Bayes via Transformerscited by 11
  6. 2024Token-based Decision Criteria Are Suboptimal in In-context Learningcited by 9
  7. 2024Leveraging Large Language Models for Wireless Symbol Detection via In-Context Learningcited by 9
  8. 2025Predicting the Performance of Black-box LLMs through Follow-up Queriescited by 9
  9. 2025LPASS: Linear Probes as Stepping Stones for vulnerability detection using compressed LLMscited by 4
  10. 2024From Unstructured Data to In-Context Learning: Exploring What Tasks Can Be Learned and Whencited by 3
  11. 2024Logistic Regression makes small LLMs strong and explainable "tens-of-shot" classifierscited by 3
  12. 2025StaICC: Standardized Evaluation for Classification Task in In-context Learningcited by 2
4 more
  1. 2025Surprise Calibration for Better In-Context Learningcited by 2
  2. 2025Stability of In-Context Learning: A Spectral Coverage Perspectivecited by 2
  3. 2025Exploring Imbalanced Annotations for Effective In-Context Learningcited by 0
  4. 2026Beyond Post-Hoc Temperature Scaling: Bilevel Optimization for LLM Calibrationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023A Study on the Calibration of In-context Learning
  2. 2024Benchmarking General-Purpose In-Context Learning
  3. 2024In-Context Language Learning: Architectures and Algorithms
  4. 2022Self-Generated In-Context Learning: Leveraging Auto-regressive Language Models as a Demonstration Generator
  5. 2024Does In-Context Learning Really Learn? Rethinking How Large Language Models Respond and Solve Tasks via In-Context Learning
  6. 2024Revisiting In-context Learning Inference Circuit in Large Language Models
  7. 2025Beyond In-Context Learning: Aligning Long-form Generation of Large Language Models via Task-Inherent Attribute Guidelines
  8. 2025Mechanistic Fine-tuning for In-context Learning