How Far Can In-Context Alignment Go? Exploring the State of In-Context Alignment

• URL: http://arxiv.org/abs/2406.11474v1
• Date: Mon, 17 Jun 2024 12:38:48 GMT
• Title: How Far Can In-Context Alignment Go? Exploring the State of In-Context Alignment
• Authors: Heyan Huang, Yinghao Li, Huashan Sun, Yu Bai, Yang Gao,
• Abstract summary: In-Context Learning (ICL) can align Large Language Models with human preferences known as In-Context Alignment (ICA) We divide context text into three categories: format, system prompt, and example. Our findings indicate that the example part is crucial for enhancing the model's alignment capabilities.
• Score: 48.0254056812898
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent studies have demonstrated that In-Context Learning (ICL), through the use of specific demonstrations, can align Large Language Models (LLMs) with human preferences known as In-Context Alignment (ICA), indicating that models can comprehend human instructions without requiring parameter adjustments. However, the exploration of the mechanism and applicability of ICA remains limited. In this paper, we begin by dividing the context text used in ICA into three categories: format, system prompt, and example. Through ablation experiments, we investigate the effectiveness of each part in enabling ICA to function effectively. We then examine how variants in these parts impact the model's alignment performance. Our findings indicate that the example part is crucial for enhancing the model's alignment capabilities, with changes in examples significantly affecting alignment performance. We also conduct a comprehensive evaluation of ICA's zero-shot capabilities in various alignment tasks. The results indicate that compared to parameter fine-tuning methods, ICA demonstrates superior performance in knowledge-based tasks and tool-use tasks. However, it still exhibits certain limitations in areas such as multi-turn dialogues and instruction following.

Related papers

• DETAIL: Task DEmonsTration Attribution for Interpretable In-context Learning [75.68193159293425]
  In-context learning (ICL) allows transformer-based language models to learn a specific task with a few "task demonstrations" without updating their parameters. We propose an influence function-based attribution technique, DETAIL, that addresses the specific characteristics of ICL. We experimentally prove the wide applicability of DETAIL by showing our attribution scores obtained on white-box models are transferable to black-box models in improving model performance.
  arXiv  Detail & Related papers  (2024-05-22T15:52:52Z)
• Bayesian Example Selection Improves In-Context Learning for Speech, Text, and Visual Modalities [15.931776592470895]
  Large language models (LLMs) can adapt to new tasks through in-context learning (ICL) This paper proposes a novel Bayesian in-Context example Selection method (ByCS) for ICL.
  arXiv  Detail & Related papers  (2024-04-23T03:42:48Z)
• ParaICL: Towards Robust Parallel In-Context Learning [74.38022919598443]
  Large language models (LLMs) have become the norm in natural language processing. Few-shot in-context learning (ICL) relies on the choice of few-shot demonstration examples. We propose a novel method named parallel in-context learning (ParaICL)
  arXiv  Detail & Related papers  (2024-03-31T05:56:15Z)
• C-ICL: Contrastive In-context Learning for Information Extraction [54.39470114243744]
  c-ICL is a novel few-shot technique that leverages both correct and incorrect sample constructions to create in-context learning demonstrations. Our experiments on various datasets indicate that c-ICL outperforms previous few-shot in-context learning methods.
  arXiv  Detail & Related papers  (2024-02-17T11:28:08Z)
• Revisiting Demonstration Selection Strategies in In-Context Learning [66.11652803887284]
  Large language models (LLMs) have shown an impressive ability to perform a wide range of tasks using in-context learning (ICL) In this work, we first revisit the factors contributing to this variance from both data and model aspects, and find that the choice of demonstration is both data- and model-dependent. We propose a data- and model-dependent demonstration selection method, textbfTopK + ConE, based on the assumption that textitthe performance of a demonstration positively correlates with its contribution to the model's understanding of the test samples.
  arXiv  Detail & Related papers  (2024-01-22T16:25:27Z)
• Towards Best Practices of Activation Patching in Language Models: Metrics and Methods [9.121998462494533]
  We examine the impact of methodological details in activation patching, including evaluation metrics and corruption methods. Backed by empirical observations, we give conceptual arguments for why certain metrics or methods may be preferred.
  arXiv  Detail & Related papers  (2023-09-27T21:53:56Z)
• In-Context Demonstration Selection with Cross Entropy Difference [95.21947716378641]
  Large language models (LLMs) can use in-context demonstrations to improve performance on zero-shot tasks. We present a cross-entropy difference (CED) method for selecting in-context demonstrations.
  arXiv  Detail & Related papers  (2023-05-24T05:04:00Z)
• In-Context Probing: Toward Building Robust Classifiers via Probing Large Language Models [5.5089506884366735]
  In this paper, we propose an alternative approach, which we term In-Context Probing (ICP) Similar to in-context learning, we contextualize the representation of the input with an instruction, but instead of decoding the output prediction, we probe the contextualized representation to predict the label. We show that ICP performs competitive or superior to finetuning and can be particularly helpful to build classifiers on top of smaller models.
  arXiv  Detail & Related papers  (2023-05-23T15:43:04Z)
• Compositional Exemplars for In-context Learning [21.961094715261133]
  Large pretrained language models (LMs) have shown impressive In-Context Learning (ICL) ability. We propose CEIL (Compositional Exemplars for In-context Learning) to model the interaction between the given input and in-context examples. We validate CEIL on 12 classification and generation datasets from 7 distinct NLP tasks, including sentiment analysis, paraphrase detection, natural language inference, commonsense reasoning, open-domain question answering, code generation, and semantic parsing.
  arXiv  Detail & Related papers  (2023-02-11T14:02:08Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.