Transformer-based Causal Language Models Perform Clustering

• URL: http://arxiv.org/abs/2402.12151v2
• Date: Sun, 3 Mar 2024 20:06:24 GMT
• Title: Transformer-based Causal Language Models Perform Clustering
• Authors: Xinbo Wu, Lav R. Varshney
• Abstract summary: We introduce a simplified instruction-following task and use synthetic datasets to analyze a Transformer-based causal language model. Our findings suggest that the model learns task-specific information by clustering data within its hidden space, with this clustering process evolving dynamically during learning.
• Score: 20.430255724239448
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Even though large language models (LLMs) have demonstrated remarkable capability in solving various natural language tasks, the capability of an LLM to follow human instructions is still a concern. Recent works have shown great improvements in the instruction-following capability via additional training for instruction-following tasks. However, the mechanisms responsible for effective instruction-following capabilities remain inadequately understood. Here, we introduce a simplified instruction-following task and use synthetic datasets to analyze a Transformer-based causal language model. Our findings suggest that the model learns task-specific information by clustering data within its hidden space, with this clustering process evolving dynamically during learning. We also demonstrate how this phenomenon assists the model in handling unseen instances, and validate our results in a more realistic setting. Furthermore, we present inspired applications regarding pre-training and alignment.

Related papers

• Learning Task Representations from In-Context Learning [73.72066284711462]
  Large language models (LLMs) have demonstrated remarkable proficiency in in-context learning. We introduce an automated formulation for encoding task information in ICL prompts as a function of attention heads. We show that our method's effectiveness stems from aligning the distribution of the last hidden state with that of an optimally performing in-context-learned model.
  arXiv  Detail & Related papers  (2025-02-08T00:16:44Z)
• An Analysis for Reasoning Bias of Language Models with Small Initialization [8.380004565348619]
  Large Language Models (LLMs) have revolutionized Natural Language Processing by demonstrating exceptional performance across diverse tasks. This study investigates the impact of the parameter initialization scale on the training behavior and task preferences of LLMs.
  arXiv  Detail & Related papers  (2025-02-05T15:23:26Z)
• The Inherent Limits of Pretrained LLMs: The Unexpected Convergence of Instruction Tuning and In-Context Learning Capabilities [51.594836904623534]
  We investigate whether instruction-tuned models possess fundamentally different capabilities from base models that are prompted using in-context examples. We show that the performance of instruction-tuned models is significantly correlated with the in-context performance of their base counterparts. Specifically, we extend this understanding to instruction-tuned models, suggesting that their pretraining data similarly sets a limiting boundary on the tasks they can solve.
  arXiv  Detail & Related papers  (2025-01-15T10:57:55Z)
• Dynamic Skill Adaptation for Large Language Models [78.31322532135272]
  We present Dynamic Skill Adaptation (DSA), an adaptive and dynamic framework to adapt novel and complex skills to Large Language Models (LLMs) For every skill, we utilize LLMs to generate both textbook-like data which contains detailed descriptions of skills for pre-training and exercise-like data which targets at explicitly utilizing the skills to solve problems for instruction-tuning. Experiments on large language models such as LLAMA and Mistral demonstrate the effectiveness of our proposed methods in adapting math reasoning skills and social study skills.
  arXiv  Detail & Related papers  (2024-12-26T22:04:23Z)
• Layer by Layer: Uncovering Where Multi-Task Learning Happens in Instruction-Tuned Large Language Models [22.676688441884465]
  Fine-tuning pre-trained large language models (LLMs) on a diverse array of tasks has become a common approach for building models. This study investigates the task-specific information encoded in pre-trained LLMs and the effects of instruction tuning on their representations.
  arXiv  Detail & Related papers  (2024-10-25T23:38:28Z)
• SwitchCIT: Switching for Continual Instruction Tuning [14.085371250265224]
  Large language models (LLMs) and multimodal models (MMs) have exhibited impressive capabilities in various domains. Continual instruction tuning is crucial to adapt a large model to evolving tasks and domains. This work addresses the catastrophic forgetting in continual instruction learning through a mechanism for routing computations to parameter-efficient tuned models.
  arXiv  Detail & Related papers  (2024-07-16T14:37:33Z)
• 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)
• Instruction Position Matters in Sequence Generation with Large Language Models [67.87516654892343]
  Large language models (LLMs) are capable of performing conditional sequence generation tasks, such as translation or summarization. We propose enhancing the instruction-following capability of LLMs by shifting the position of task instructions after the input sentences.
  arXiv  Detail & Related papers  (2023-08-23T12:36:57Z)
• SINC: Self-Supervised In-Context Learning for Vision-Language Tasks [64.44336003123102]
  We propose a framework to enable in-context learning in large language models. A meta-model can learn on self-supervised prompts consisting of tailored demonstrations. Experiments show that SINC outperforms gradient-based methods in various vision-language tasks.
  arXiv  Detail & Related papers  (2023-07-15T08:33:08Z)
• Concept-aware Training Improves In-context Learning Ability of Language Models [0.0]
  Many recent language models (LMs) of Transformers family exhibit so-called in-context learning (ICL) ability. We propose a method to create LMs able to better utilize the in-context information. We measure that data sampling of Concept-aware Training consistently improves models' reasoning ability.
  arXiv  Detail & Related papers  (2023-05-23T07:44:52Z)

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.