Parallel Structures in Pre-training Data Yield In-Context Learning (2024.acl-long)
Copied to clipboard
| Challenge: | Pre-trained language models (LMs) are capable of in-context learning (ICL) however, it is unclear where this ability comes from as there is a stark distribution shift between pre-training text and ICL prompts. |
| Approach: | They find that pre-trained language models are capable of in-context learning (ICL) they detect parallel structures in the pre-training data and conduct ablation experiments to study their effect on ICL. |
| Outcome: | The proposed model can adapt to a task with a few examples given in the prompt without any parameter update. |
Similar Papers
Understanding In-Context Learning via Supportive Pretraining Data (2023.acl-long)
Copied to clipboard
| Challenge: | In-context learning (ICL) is a form of learning that provides a handful of examples at inference time, but it is not well understood why it emerges as the model has never been specifically trained on such demonstrations. |
| Approach: | They adapt an iterative, gradient-based approach to find a small subset of pretraining data that supports ICL and compare it with random subsets of pretrain data. |
| Outcome: | The proposed method improves the model's ICL ability by 18% if it is continued on a small subset of pretraining data. |
What Do Language Models Learn in Context? The Structured Task Hypothesis. (2024.acl-long)
Copied to clipboard
| Challenge: | Pre-trained large language models have exhibited an impressive ability to learn in context across various domains, e.g., code generation, education, medicine and even medicine. |
| Approach: | They taxonomize existing candidate theories into three competing hypotheses that explain LLMs’ ability to learn in context. |
| Outcome: | The proposed model can learn a task from in-context examples presented in a demonstration and generalize it to the prompt. |
In-context Learning Generalizes, But Not Always Robustly: The Case of Syntax (2024.naacl-long)
Copied to clipboard
| Challenge: | In-context learning is a common method for teaching large language models new tasks . given labeled examples in the input context, the model learns to perform the task without weight updates. |
| Approach: | They examine whether models guided via ICL infer the underlying structure of the task defined by the context or rely on superficial heuristics that only generalize to identically distributed examples. |
| Outcome: | The proposed model generalizes syntactically or linearly on out-of-distribution examples . the proposed model is able to generalize better on pre-trained models . |
On the In-context Generation of Language Models (2024.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) have the ability of in-context generation (ICG) when given an in-text prompt, they can implicitly recognize the pattern of the examples and complete the prompt in the desired way. |
| Approach: | They propose a plausible latent variable model to model the distribution of pretrained corpora and formalize ICG as a problem of next topic prediction. |
| Outcome: | The proposed model can model the distribution of pretrained corpora and then formalize ICG as a problem of next topic prediction. |
What In-Context Learning “Learns” In-Context: Disentangling Task Recognition and Task Learning (2023.findings-acl)
Copied to clipboard
| Challenge: | Large language models (LLMs) can perform in-context learning (ICL) with only a few demonstrations, but its mechanisms are not well-understood. |
| Approach: | They characterize two ways in which LLMs leverage demonstrations to solve tasks with a few demonstrations. |
| Outcome: | The proposed model achieves non-trivial performance with only TR, and TR does not improve with larger models or more demonstrations. |
In-Context Learning Creates Task Vectors (2023.findings-emnlp)
Copied to clipboard
| Challenge: | In-context learning (ICL) is a powerful new learning paradigm for Large Language Models (LLMs). |
| Approach: | They propose to use a model with a prompt and a query to learn a mapping based on two examples to produce the output. |
| Outcome: | The proposed model can learn functions from a simple structure based on a training set and a single task vector calculated from the training set. |
An Empirical Study of In-context Learning in LLMs for Machine Translation (2024.findings-acl)
Copied to clipboard
| Challenge: | Recent studies focus on optimizing translation quality, with limited attention to understanding specific aspects of ICL that influence the said quality. |
| Approach: | They conduct the first of its kind, exhaustive study of in-context learning for machine translation (MT) they establish that ICL is primarily example-driven and not instruction-driven . |
| Outcome: | The proposed model is based on examples and not instruction-driven learning. |
A Survey to Recent Progress Towards Understanding In-Context Learning (2025.findings-naacl)
Copied to clipboard
| Challenge: | Existing research on In-Context Learning (ICL) is unclear, despite empirical success . a data generation perspective is used to interpret ICL . |
| Approach: | They propose to use data generation to reinterpret recent efforts from a systematic angle to demonstrate the potential broader usage of ICL. |
| Outcome: | The proposed model can learn from examples provided in the prompt, enabling downstream generalization without the need for gradient updates. |
ICLEval: Evaluating In-Context Learning Ability of Large Language Models (2025.coling-main)
Copied to clipboard
| Challenge: | Existing evaluation frameworks focus on language abilities and knowledge, often overlooking the assessment of ICL ability. |
| Approach: | They propose to evaluate the ICL ability of Large Language Models (LLMs) using the ICLEval benchmark. |
| Outcome: | The proposed benchmark demonstrates that ICL ability is universally present in different LLMs and model size is not the sole determinant of ICL efficacy. |
Deciphering the Impact of Pretraining Data on Large Language Models through Machine Unlearning (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing studies have suggested that the composition of the pretraining corpus exerts a significant impact upon the performance of LLMs. |
| Approach: | They analyze the impact of 48 datasets from 5 major categories of pretraining data of Large Language Models and measure their impacts on LLMs using benchmarks about nine major categories. |
| Outcome: | The proposed analysis provides insights into the organization of data to support more efficient pretraining of Large Language Models. |