Skip to content
AI News HubLIVE
Original source2 min read

Speaking in Self-Assessing Tongues: On the Verbalized Confidence of LLMs in Machine Translation

Summary

A new paper investigates the reliability of large language models' confidence in their machine translation outputs. The authors propose five verbalized methods to extract per-token confidence without relying on internal signals, and compare them with internal certainty signals in terms of fine-grained error detection and calibration. Results show similar overall performance but variation across models, and surprisingly little to no correlation between internal and verbalized methods.

SourcearXiv Computational LinguisticsAuthor: Ali Marashian, Alexis Palmer, Katharina von der Wense
Speaking in Self-Assessing Tongues: On the Verbalized Confidence of LLMs in Machine Translation
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[2606.17234] Speaking in Self-Assessing Tongues: On the Verbalized Confidence of LLMs in Machine Translation

[Submitted on 15 Jun 2026]

Title:Speaking in Self-Assessing Tongues: On the Verbalized Confidence of LLMs in Machine Translation

View a PDF of the paper titled Speaking in Self-Assessing Tongues: On the Verbalized Confidence of LLMs in Machine Translation, by Ali Marashian and 2 other authors

View PDF HTML (experimental)

Abstract:The rapid rise in popularity of large language models (LLMs) for translation calls for a thorough study of the reliability of their confidence in their own outputs. Unlike many generation tasks, translation errors and confidence levels can be useful at different levels of granularity (tokens, words, or spans). Unsupervised approaches based on internal signals like predicted probabilities can be misleading because they reflect certainty among alternatives rather than correctness. In addition, they require access to such internal signals. Here, we devise five verbalized methods of extracting an LLM's per-token confidence without those shortcomings and compare their reliability with that of the model's internal signals of certainty. We evaluate reliability using two forms of alignment: fine-grained error detection and calibration. For both, internal and verbalized methods perform similarly, although results vary by model. Interestingly, we find little to no correlation between internal and verbalized methods.

Subjects:

Computation and Language (cs.CL)

Cite as: arXiv:2606.17234 [cs.CL]

(or arXiv:2606.17234v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2606.17234

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Ali Marashian [view email] [v1] Mon, 15 Jun 2026 19:27:47 UTC (2,830 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Speaking in Self-Assessing Tongues: On the Verbalized Confidence of LLMs in Machine Translation, by Ali Marashian and 2 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CL

new | recent | 2026-06

Change to browse by:

cs

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

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?)

Key points and analysis

Article intelligence

EngineersAdvanced

Key points

  • Five verbalized methods extract per-token confidence without internal signal access
  • Internal and verbalized methods perform similarly in error detection and calibration, but vary by model
  • Little to no correlation found between internal and verbalized confidence
  • Unsupervised internal-signal approaches can be misleading as they reflect certainty among alternatives, not correctness

Highlights and analysis are generated automatically and may contain errors. Check the original source.