Training Dynamics of Parametric and In-Context Knowledge Utilization in Language Models

• URL: http://arxiv.org/abs/2510.02370v1
• Date: Mon, 29 Sep 2025 06:18:18 GMT
• Title: Training Dynamics of Parametric and In-Context Knowledge Utilization in Language Models
• Authors: Minsung Kim, Dong-Kyum Kim, Jea Kwon, Nakyeong Yang, Kyomin Jung, Meeyoung Cha,
• Abstract summary: Large language models often encounter conflicts between in-context knowledge retrieved at inference time and parametric knowledge acquired during pretraining.<n>We present the first controlled study of how training conditions influence models' use of in-context and parametric knowledge.<n>Our experiments reveal that intra-document repetition of facts fosters the development of both parametric and in-context capabilities.
• Score: 31.829376135133554
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Large language models often encounter conflicts between in-context knowledge retrieved at inference time and parametric knowledge acquired during pretraining. Models that accept external knowledge uncritically are vulnerable to misinformation, whereas models that adhere rigidly to parametric knowledge fail to benefit from retrieval. Despite the widespread adoption of retrieval-augmented generation, we still lack a systematic understanding of what shapes knowledge-arbitration strategies during training. This gap risks producing pretrained models with undesirable arbitration behaviors and, consequently, wasting substantial computational resources after the pretraining budget has already been spent. To address this problem, we present the first controlled study of how training conditions influence models' use of in-context and parametric knowledge, and how they arbitrate between them. We train transformer-based language models on a synthetic biographies corpus while systematically controlling various conditions. Our experiments reveal that intra-document repetition of facts fosters the development of both parametric and in-context capabilities. Moreover, training on a corpus that contains inconsistent information or distributional skew encourages models to develop robust strategies for leveraging parametric and in-context knowledge. Rather than viewing these non-ideal properties as artifacts to remove, our results indicate that they are important for learning robust arbitration. These insights offer concrete, empirical guidance for pretraining models that harmoniously integrate parametric and in-context knowledge.

Related papers

• Auditing Language Model Unlearning via Information Decomposition [68.48660428111593]
  We introduce an interpretable, information-theoretic framework for auditing unlearning using Partial Information Decomposition (PID)<n>By comparing model representations before and after unlearning, we decompose the mutual information with the forgotten data into distinct components, formalizing the notions of unlearned and residual knowledge.<n>Our work introduces a principled, representation-level audit for unlearning, offering theoretical insight and actionable tools for safer deployment of language models.
  arXiv  Detail & Related papers  (2026-01-21T15:51:19Z)
• Where Knowledge Collides: A Mechanistic Study of Intra-Memory Knowledge Conflict in Language Models [8.965740058804197]
  In language models (LMs) intra-memory knowledge conflict largely arises when inconsistent information about the same event is encoded within the model's parametric knowledge.<n>We use mechanistic interpretability methods to identify where and how conflicting knowledge from pre-training data is encoded within LMs.<n>Our findings contribute to a growing body of evidence that specific internal components of a language model are responsible for encoding conflicting knowledge from pre-training.
  arXiv  Detail & Related papers  (2026-01-14T12:45:52Z)
• FaithfulRAG: Fact-Level Conflict Modeling for Context-Faithful Retrieval-Augmented Generation [37.28571879699906]
  Large language models (LLMs) augmented with retrieval systems have demonstrated significant potential in handling knowledge-intensive tasks.<n>This paper proposes FaithfulRAG, a novel framework that resolves knowledge conflicts by explicitly modeling discrepancies between the models parametric knowledge and retrieved context.
  arXiv  Detail & Related papers  (2025-06-10T16:02:54Z)
• Task Matters: Knowledge Requirements Shape LLM Responses to Context-Memory Conflict [13.091464232666835]
  Large Language Models require both contextual knowledge and parametric memory, but these sources can disagree.<n>We study this question with a model-agnostic diagnostic framework that automatically detects disagreements between a model's beliefs and a curated knowledge set.<n>We find that performance degradation from conflict correlates with a task's knowledge reliance.
  arXiv  Detail & Related papers  (2025-06-06T19:20:23Z)
• Mitigating Knowledge Conflicts in Language Model-Driven Question Answering [15.29366851382021]
  Two fundamental knowledge sources play crucial roles in document-based question answering and document summarization systems.<n>Recent studies revealed a significant challenge: when there exists a misalignment between the model's inherent knowledge and the ground truth answers in training data, the system may exhibit problematic behaviors during inference.<n>Our investigation proposes a strategy to minimize hallucination by building explicit connection between source inputs and generated outputs.
  arXiv  Detail & Related papers  (2024-11-18T07:33:10Z)
• Context-Parametric Inversion: Why Instruction Finetuning Can Worsen Context Reliance [68.56701216210617]
  In-principle, one would expect models to adapt to the user context better after instruction finetuning.<n>We observe a surprising failure mode: during instruction tuning, the context reliance under knowledge conflicts initially increases as expected, but then gradually decreases.
  arXiv  Detail & Related papers  (2024-10-14T17:57:09Z)
• The KITMUS Test: Evaluating Knowledge Integration from Multiple Sources in Natural Language Understanding Systems [87.3207729953778]
  We evaluate state-of-the-art coreference resolution models on our dataset. Several models struggle to reason on-the-fly over knowledge observed both at pretrain time and at inference time. Still, even the best performing models seem to have difficulties with reliably integrating knowledge presented only at inference time.
  arXiv  Detail & Related papers  (2022-12-15T23:26:54Z)
• Large Language Models with Controllable Working Memory [64.71038763708161]
  Large language models (LLMs) have led to a series of breakthroughs in natural language processing (NLP) What further sets these models apart is the massive amounts of world knowledge they internalize during pretraining. How the model's world knowledge interacts with the factual information presented in the context remains under explored.
  arXiv  Detail & Related papers  (2022-11-09T18:58:29Z)
• Does Pre-training Induce Systematic Inference? How Masked Language Models Acquire Commonsense Knowledge [91.15301779076187]
  We introduce verbalized knowledge into the minibatches of a BERT model during pre-training and evaluate how well the model generalizes to supported inferences. We find generalization does not improve over the course of pre-training, suggesting that commonsense knowledge is acquired from surface-level, co-occurrence patterns rather than induced, systematic reasoning.
  arXiv  Detail & Related papers  (2021-12-16T03:13:04Z)
• Precise Tradeoffs in Adversarial Training for Linear Regression [55.764306209771405]
  We provide a precise and comprehensive understanding of the role of adversarial training in the context of linear regression with Gaussian features. We precisely characterize the standard/robust accuracy and the corresponding tradeoff achieved by a contemporary mini-max adversarial training approach. Our theory for adversarial training algorithms also facilitates the rigorous study of how a variety of factors (size and quality of training data, model overparametrization etc.) affect the tradeoff between these two competing accuracies.
  arXiv  Detail & Related papers  (2020-02-24T19:01:47Z)

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.