LMs as Task-Specific Knowledge Bases: An Interpretability Analysis

• URL: http://arxiv.org/abs/2606.27237v1
• Date: Thu, 25 Jun 2026 16:22:43 GMT
• Title: LMs as Task-Specific Knowledge Bases: An Interpretability Analysis
• Abstract summary: We investigate whether language models encode knowledge in a task-specific manner.<n>We show that facts acquired on one task frequently fail to co-emerge on others during training.<n>We show that chain-of-thought reasoning draws part of its effectiveness from engaging task-specific parameters beyond those tied to the evaluation task.
• Score: 45.933429785602506
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Language models (LMs) capture large amounts of factual knowledge applicable to a wide range of tasks, motivating the view of their parameters as a knowledge base. An important property of knowledge bases is that different queries for the same fact return consistent results, drawing on a single source of truth. We investigate whether LMs satisfy this property through behavioral and mechanistic analyses. Our results suggest that they encode knowledge in a task-specific manner. Behaviorally, facts acquired on one task frequently fail to co-emerge on others during training. Parameter localization experiments suggest a mechanistic explanation, revealing distinct parameter subsets underlying different tasks for the same fact. Finally, we show that chain-of-thought reasoning draws part of its effectiveness from engaging task-specific parameters beyond those tied to the evaluation task. Our findings suggest that what the model knows and how it is asked are intertwined in parameter space, undermining the "knowledge base" analogy and carrying implications for the reliability and controllability of factual knowledge in LMs.

Related papers

• 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)
• Latent Causal Probing: A Formal Perspective on Probing with Causal Models of Data [3.376269351435396]
  We develop a formal perspective on probing using structural causal models (SCM) We extend a recent study of LMs in the context of a synthetic grid-world navigation task. Our techniques provide robust empirical evidence for the ability of LMs to induce the latent concepts underlying text.
  arXiv  Detail & Related papers  (2024-07-18T17:59:27Z)
• A Unified Causal View of Instruction Tuning [76.1000380429553]
  We develop a meta Structural Causal Model (meta-SCM) to integrate different NLP tasks under a single causal structure of the data. Key idea is to learn task-required causal factors and only use those to make predictions for a given task.
  arXiv  Detail & Related papers  (2024-02-09T07:12:56Z)
• Seeking Neural Nuggets: Knowledge Transfer in Large Language Models from a Parametric Perspective [106.92016199403042]
  We empirically investigate knowledge transfer from larger to smaller models through a parametric perspective. We employ sensitivity-based techniques to extract and align knowledge-specific parameters between different large language models. Our findings highlight the critical factors contributing to the process of parametric knowledge transfer.
  arXiv  Detail & Related papers  (2023-10-17T17:58:34Z)
• Pre-training Language Models with Deterministic Factual Knowledge [42.812774794720895]
  We propose to let PLMs learn the deterministic relationship between the remaining context and the masked content. Two pre-training tasks are introduced to motivate PLMs to rely on the deterministic relationship when filling masks. Experiments indicate that the continuously pre-trained PLMs achieve better robustness in factual knowledge capturing.
  arXiv  Detail & Related papers  (2022-10-20T11:04:09Z)
• Distribution Matching for Heterogeneous Multi-Task Learning: a Large-scale Face Study [75.42182503265056]
  Multi-Task Learning has emerged as a methodology in which multiple tasks are jointly learned by a shared learning algorithm. We deal with heterogeneous MTL, simultaneously addressing detection, classification & regression problems. We build FaceBehaviorNet, the first framework for large-scale face analysis, by jointly learning all facial behavior tasks.
  arXiv  Detail & Related papers  (2021-05-08T22:26:52Z)
• Knowledge-driven Data Construction for Zero-shot Evaluation in Commonsense Question Answering [80.60605604261416]
  We propose a novel neuro-symbolic framework for zero-shot question answering across commonsense tasks. We vary the set of language models, training regimes, knowledge sources, and data generation strategies, and measure their impact across tasks. We show that, while an individual knowledge graph is better suited for specific tasks, a global knowledge graph brings consistent gains across different tasks.
  arXiv  Detail & Related papers  (2020-11-07T22:52:21Z)
• oLMpics -- On what Language Model Pre-training Captures [84.60594612120173]
  We propose eight reasoning tasks, which require operations such as comparison, conjunction, and composition. A fundamental challenge is to understand whether the performance of a LM on a task should be attributed to the pre-trained representations or to the process of fine-tuning on the task data.
  arXiv  Detail & Related papers  (2019-12-31T12:11:35Z)

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.