Papers
Topics
Authors
Recent
Detailed Answer
Quick Answer
Concise responses based on abstracts only
Detailed Answer
Well-researched responses based on abstracts and relevant paper content.
Custom Instructions Pro
Preferences or requirements that you'd like Emergent Mind to consider when generating responses
Gemini 2.5 Flash
Gemini 2.5 Flash 54 tok/s
Gemini 2.5 Pro 50 tok/s Pro
GPT-5 Medium 18 tok/s Pro
GPT-5 High 31 tok/s Pro
GPT-4o 105 tok/s Pro
Kimi K2 182 tok/s Pro
GPT OSS 120B 466 tok/s Pro
Claude Sonnet 4 40 tok/s Pro
2000 character limit reached

Mind the instructions: a holistic evaluation of consistency and interactions in prompt-based learning (2310.13486v1)

Published 20 Oct 2023 in cs.CL and cs.AI

Abstract: Finding the best way of adapting pre-trained LLMs to a task is a big challenge in current NLP. Just like the previous generation of task-tuned models (TT), models that are adapted to tasks via in-context-learning (ICL) are robust in some setups but not in others. Here, we present a detailed analysis of which design choices cause instabilities and inconsistencies in LLM predictions. First, we show how spurious correlations between input distributions and labels -- a known issue in TT models -- form only a minor problem for prompted models. Then, we engage in a systematic, holistic evaluation of different factors that have been found to influence predictions in a prompting setup. We test all possible combinations of a range of factors on both vanilla and instruction-tuned (IT) LLMs of different scale and statistically analyse the results to show which factors are the most influential, interactive or stable. Our results show which factors can be used without precautions and which should be avoided or handled with care in most settings.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (49)
  1. What learning algorithm is in-context learning? investigations with linear models. ArXiv preprint, abs/2211.15661.
  2. Exploring the landscape of distributional robustness for question answering models. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 5971–5987, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
  3. PromptSource: An integrated development environment and repository for natural language prompts. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics: System Demonstrations, pages 93–104, Dublin, Ireland. Association for Computational Linguistics.
  4. A multitask, multilingual, multimodal evaluation of chatgpt on reasoning, hallucination, and interactivity. ArXiv preprint, abs/2302.04023.
  5. Beat the AI: Investigating adversarial human annotation for reading comprehension. Transactions of the Association for Computational Linguistics, 8:662–678.
  6. Language models are few-shot learners. In Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6-12, 2020, virtual.
  7. Jacob Cohen. 1960. A coefficient of agreement for nominal scales. Educational and psychological measurement, 20(1):37–46.
  8. Llm. int8 (): 8-bit matrix multiplication for transformers at scale. ArXiv preprint, abs/2208.07339.
  9. What can transformers learn in-context? a case study of simple function classes. Advances in Neural Information Processing Systems, 35:30583–30598.
  10. Shortcut learning in deep neural networks. Nature Machine Intelligence, 2(11):665–673.
  11. Are we modeling the task or the annotator? an investigation of annotator bias in natural language understanding datasets. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 1161–1166, Hong Kong, China. Association for Computational Linguistics.
  12. Annotation artifacts in natural language inference data. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers), pages 107–112, New Orleans, Louisiana. Association for Computational Linguistics.
  13. Using pre-training can improve model robustness and uncertainty. In Proceedings of the 36th International Conference on Machine Learning, ICML 2019, 9-15 June 2019, Long Beach, California, USA, volume 97 of Proceedings of Machine Learning Research, pages 2712–2721. PMLR.
  14. Pretrained transformers improve out-of-distribution robustness. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 2744–2751, Online. Association for Computational Linguistics.
  15. Lora: Low-rank adaptation of large language models. In The Tenth International Conference on Learning Representations, ICLR 2022, Virtual Event, April 25-29, 2022. OpenReview.net.
  16. A taxonomy and review of generalization research in nlp. Nature Machine Intelligence, 5:1161–1174.
  17. Robin Jia and Percy Liang. 2017. Adversarial examples for evaluating reading comprehension systems. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 2021–2031, Copenhagen, Denmark. Association for Computational Linguistics.
  18. Learning the difference that makes A difference with counterfactually-augmented data. In 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020. OpenReview.net.
  19. When choosing plausible alternatives, clever hans can be clever. In Proceedings of the First Workshop on Commonsense Inference in Natural Language Processing, pages 33–42, Hong Kong, China. Association for Computational Linguistics.
  20. Are prompt-based models clueless? In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2333–2352, Dublin, Ireland. Association for Computational Linguistics.
  21. Holistic evaluation of language models. ArXiv preprint, abs/2211.09110.
  22. Roberta: A robustly optimized bert pretraining approach. ArXiv preprint, abs/1907.11692.
  23. Fantastically ordered prompts and where to find them: Overcoming few-shot prompt order sensitivity. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 8086–8098, Dublin, Ireland. Association for Computational Linguistics.
  24. Right for the wrong reasons: Diagnosing syntactic heuristics in natural language inference. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 3428–3448, Florence, Italy. Association for Computational Linguistics.
  25. The effect of natural distribution shift on question answering models. In Proceedings of the 37th International Conference on Machine Learning, ICML 2020, 13-18 July 2020, Virtual Event, volume 119 of Proceedings of Machine Learning Research, pages 6905–6916. PMLR.
  26. Rethinking the role of demonstrations: What makes in-context learning work? In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 11048–11064, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
  27. Reframing instructional prompts to GPTk’s language. In Findings of the Association for Computational Linguistics: ACL 2022, pages 589–612, Dublin, Ireland. Association for Computational Linguistics.
  28. Few-shot fine-tuning vs. in-context learning: A fair comparison and evaluation. ArXiv preprint, abs/2305.16938.
  29. Adversarial NLI: A new benchmark for natural language understanding. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 4885–4901, Online. Association for Computational Linguistics.
  30. Timothy Niven and Hung-Yu Kao. 2019. Probing neural network comprehension of natural language arguments. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 4658–4664, Florence, Italy. Association for Computational Linguistics.
  31. Evaluating task understanding through multilingual consistency: A chatgpt case study. ArXiv preprint, abs/2305.11662.
  32. Training language models to follow instructions with human feedback. Advances in Neural Information Processing Systems, 35:27730–27744.
  33. Hypothesis only baselines in natural language inference. In Proceedings of the Seventh Joint Conference on Lexical and Computational Semantics, pages 180–191, New Orleans, Louisiana. Association for Computational Linguistics.
  34. SQuAD: 100,000+ questions for machine comprehension of text. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, pages 2383–2392, Austin, Texas. Association for Computational Linguistics.
  35. On causal and anticausal learning. In Proceedings of the 29th International Conference on Machine Learning, ICML 2012, Edinburgh, Scotland, UK, June 26 - July 1, 2012. icml.cc / Omnipress.
  36. Prompting gpt-3 to be reliable.
  37. Stanford alpaca: An instruction-following llama model. https://github.com/tatsu-lab/stanford_alpaca.
  38. Llama: Open and efficient foundation language models. ArXiv preprint, abs/2302.13971.
  39. Self-instruct: Aligning language model with self generated instructions. ArXiv preprint, abs/2212.10560.
  40. Bilateral multi-perspective matching for natural language sentences. In Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence, IJCAI 2017, Melbourne, Australia, August 19-25, 2017, pages 4144–4150. ijcai.org.
  41. Finetuned language models are zero-shot learners. In The Tenth International Conference on Learning Representations, ICLR 2022, Virtual Event, April 25-29, 2022. OpenReview.net.
  42. Larger language models do in-context learning differently. ArXiv preprint, abs/2303.03846.
  43. A broad-coverage challenge corpus for sentence understanding through inference. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers), pages 1112–1122, New Orleans, Louisiana. Association for Computational Linguistics.
  44. Huggingface’s transformers: State-of-the-art natural language processing. ArXiv preprint, abs/1910.03771.
  45. Ground-truth labels matter: A deeper look into input-label demonstrations. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 2422–2437, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
  46. PAWS: Paraphrase adversaries from word scrambling. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 1298–1308, Minneapolis, Minnesota. Association for Computational Linguistics.
  47. Calibrate before use: Improving few-shot performance of language models. In Proceedings of the 38th International Conference on Machine Learning, ICML 2021, 18-24 July 2021, Virtual Event, volume 139 of Proceedings of Machine Learning Research, pages 12697–12706. PMLR.
  48. Adapting language models for zero-shot learning by meta-tuning on dataset and prompt collections. In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 2856–2878, Punta Cana, Dominican Republic. Association for Computational Linguistics.
  49. Lima: Less is more for alignment.
Citations (19)
View on Semantic Scholar
List To Do Tasks Checklist Streamline Icon: https://streamlinehq.com

Collections

Sign up for free to add this paper to one or more collections.

User Circle Single Streamline Icon: https://streamlinehq.com Sign Up

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
Dice Question Streamline Icon: https://streamlinehq.com

Follow-Up Questions

We haven't generated follow-up questions for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now