Skip to main content
Cornell University
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
arxiv logo > cs > arXiv:2406.12673

Help | Advanced Search

arXiv logo
Cornell University Logo

quick links

  • Login
  • Help Pages
  • About

Computer Science > Computation and Language

arXiv:2406.12673 (cs)
[Submitted on 18 Jun 2024 (v1), last revised 29 Oct 2024 (this version, v2)]

Title:Estimating Knowledge in Large Language Models Without Generating a Single Token

Authors:Daniela Gottesman, Mor Geva
View a PDF of the paper titled Estimating Knowledge in Large Language Models Without Generating a Single Token, by Daniela Gottesman and Mor Geva
View PDF HTML (experimental)
Abstract:To evaluate knowledge in large language models (LLMs), current methods query the model and then evaluate its generated responses. In this work, we ask whether evaluation can be done before the model has generated any text. Concretely, is it possible to estimate how knowledgeable a model is about a certain entity, only from its internal computation? We study this question with two tasks: given a subject entity, the goal is to predict (a) the ability of the model to answer common questions about the entity, and (b) the factuality of open-ended responses generated by the model about the entity. Experiments with a variety of LLMs show that KEEN, a simple probe trained over internal subject representations, succeeds at both tasks - correlating with both the QA accuracy of the model per-subject and FActScore, a recent factuality metric in open-ended generation. Moreover, KEEN naturally aligns with the model's hedging behavior and faithfully reflects changes in the model's knowledge after fine-tuning. Lastly, we show a more interpretable yet equally performant variant of KEEN, which highlights a small set of tokens indicative of clusters and gaps in the model's knowledge. Being simple and lightweight, KEEN can be leveraged to guide decisions such as when it is appropriate to apply further training or augment queries with retrieval.
Comments: Accepted at EMNLP 2024 Main Conference
Subjects: Computation and Language (cs.CL)
Cite as: arXiv:2406.12673 [cs.CL]
  (or arXiv:2406.12673v2 [cs.CL] for this version)
  https://doi.org/10.48550/arXiv.2406.12673
arXiv-issued DOI via DataCite

Submission history

From: Daniela Gottesman [view email]
[v1] Tue, 18 Jun 2024 14:45:50 UTC (10,130 KB)
[v2] Tue, 29 Oct 2024 08:40:25 UTC (10,130 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Estimating Knowledge in Large Language Models Without Generating a Single Token, by Daniela Gottesman and Mor Geva
  • View PDF
  • HTML (experimental)
  • TeX Source
  • Other Formats
license icon view license
Current browse context:
cs.CL
< prev   |   next >
new | recent | 2024-06
Change to browse by:
cs

References & Citations

  • NASA ADS
  • Google Scholar
  • Semantic Scholar
a export BibTeX citation Loading...

BibTeX formatted citation

×
Data provided by:

Bookmark

BibSonomy logo Reddit logo

Bibliographic and Citation Tools

Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)

Code, Data and Media Associated with this Article

alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)

Demos

Replicate (What is Replicate?)
Hugging Face Spaces (What is Spaces?)
TXYZ.AI (What is TXYZ.AI?)

Recommenders and Search Tools

Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
  • Author
  • Venue
  • Institution
  • Topic

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)
  • About
  • Help
  • contact arXivClick here to contact arXiv Contact
  • subscribe to arXiv mailingsClick here to subscribe Subscribe
  • Copyright
  • Privacy Policy
  • Web Accessibility Assistance
  • arXiv Operational Status
    Get status notifications via email or slack