mlsanityData to 14 Oct 2026

Experimental Design for Active Transductive Inference in Large Language Models

arXiv 2404.08846 ↗First posted 12 Apr 2024cs.LG cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

One emergent ability of large language models (LLMs) is that query-specific examples can be included in the prompt at inference time. In this work, we use active learning for adaptive prompt design and call it Active In-context Prompt Design (AIPD). We design the LLM prompt by adaptively choosing few-shot examples from a training set to optimize performance on a test set. The training examples are initially unlabeled and we obtain the label of the most informative ones, which maximally reduces uncertainty in the LLM prediction. We propose two algorithms, GO and SAL, which differ in how the few-shot examples are chosen. We analyze these algorithms in linear models: first GO and then use its equivalence with SAL. We experiment with many different tasks in small, medium-sized, and large language models; and show that GO and SAL outperform other methods for choosing few-shot examples in the LLM prompt at inference time.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 57 of its 91 references that are in the map. First posted on arXiv 12 Apr 2024.

Builds on

57 of 91 references.

References · 57

  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. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  4. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  5. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,795
  6. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,780
  7. 2023Mistral 7Bcited by 3,164
  8. 2021Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processingcited by 2,234
  9. 2022Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language modelscited by 1,998
  10. 2022Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Themcited by 1,807
  11. 2022Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?cited by 1,434
  12. 2021An Explanation of In-context Learning as Implicit Bayesian Inferencecited by 938
38 more
  1. 2023A Survey on Multimodal Large Language Modelscited by 909
  2. 2017Deep Bayesian Active Learning with Image Datacited by 819
  3. 2023The RefinedWeb Dataset for Falcon LLM: Outperforming Curated Corpora with Web Data, and Web Data Onlycited by 763
  4. 2022Automatic Chain of Thought Prompting in Large Language Modelscited by 688
  5. 2019Deep Batch Active Learning by Diverse, Uncertain Gradient Lower Boundscited by 610
  6. 2020A Survey of Deep Active Learningcited by 554
  7. 2014OpenML: networked science in machine learningcited by 526
  8. 2019BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learningcited by 432
  9. 2023Hard Prompts Made Easy: Gradient-Based Discrete Optimization for Prompt Tuning and Discoverycited by 336
  10. 2022One Embedder, Any Task: Instruction-Finetuned Text Embeddingscited by 333
  11. 2023Transformers as Statisticians: Provable In-Context Learning with In-Context Algorithm Selectioncited by 298
  12. 2019Compositionality decomposed: how do neural networks generalise?cited by 265
  13. 2021Efficiently Identifying Task Groupings for Multi-Task Learningcited by 222
  14. 2023Large Language Models as General Pattern Machinescited by 197
  15. 2022Active Example Selection for In-Context Learningcited by 196
  16. 2023Active Prompting with Chain-of-Thought for Large Language Modelscited by 154
  17. 2023What Makes Good Examples for Visual In-Context Learning?cited by 150
  18. 2022LIFT: Language-Interfaced Fine-Tuning for Non-Language Machine Learning Taskscited by 148
  19. 2023Supervised Pretraining Can Learn In-Context Reinforcement Learningcited by 139
  20. 2020Cold-start Active Learning through Self-supervised Language Modelingcited by 128
  21. 2021Batch Active Learning at Scalecited by 127
  22. 2023In-Context Learning Unlocked for Diffusion Modelscited by 104
  23. 2023Align Your Prompts: Test-Time Prompting with Distribution Alignment for Zero-Shot Generalizationcited by 101
  24. 2022A Comparative Survey of Deep Active Learningcited by 89
  25. 2021Gone Fishing: Neural Active Learning with Fisher Embeddingscited by 88
  26. 2017Deep Active Learning over the Long Tailcited by 84
  27. 2023Active Instruction Tuning: Improving Cross-Task Generalization by Training on Prompt Sensitive Taskscited by 55
  28. 2023Active Learning Principles for In-Context Learning with Large Language Modelscited by 54
  29. 2022Reward-Free RL is No Harder Than Reward-Aware RL in Linear Markov Decision Processescited by 49
  30. 2021Minimax Optimal Fixed-Budget Best Arm Identification in Linear Banditscited by 32
  31. 2022Multimodal Knowledge Alignment with Reinforcement Learningcited by 30
  32. 2023Universality and Limitations of Prompt Tuningcited by 30
  33. 2015Convergence Rates of Active Learning for Maximum Likelihood Estimationcited by 25
  34. 2024An Experimental Design Framework for Label-Efficient Supervised Finetuning of Large Language Modelscited by 23
  35. 2021Fixed-Budget Best-Arm Identification in Structured Banditscited by 22
  36. 2023Efficient Detection of LLM-generated Texts with a Bayesian Surrogate Modelcited by 18
  37. 2021Nearly Optimal Algorithms for Level Set Estimationcited by 17
  38. 2023`A-la-carte Prompt Tuning (APT): Combining Distinct Data Via Composable Promptingcited by 17

50 most cited of 57 All 57 (JSON)

Built on by

Not cited yet.