How Language Models Fail: Token-Level Signatures of Committed and Persistent Reasoning Failures
This paper identifies two distinct processes of language model reasoning failures through token-level uncertainty signals: committed failure (early lock-in to incorrect paths) and persistent uncertainty (accumulating uncertainty). The framework is validated across 23 model-dataset configurations, with falsifiable predictions holding in 20 of 23 cases, and shows implications for self-consistency methods.
[2606.06635] How Language Models Fail: Token-Level Signatures of Committed and Persistent Reasoning Failures
[Submitted on 4 Jun 2026]
Title:How Language Models Fail: Token-Level Signatures of Committed and Persistent Reasoning Failures
View a PDF of the paper titled How Language Models Fail: Token-Level Signatures of Committed and Persistent Reasoning Failures, by Tanvi Thoria and 3 other authors
View PDF HTML (experimental)
Abstract:Failures in language model reasoning emerge through distinct processes that leave identifiable signatures in the reasoning trace. We characterize these failures using token-level uncertainty signals, finding they arise through two empirically distinguishable processes. The first is committed failure, in which a model locks onto an incorrect reasoning path early in its trace. A central diagnostic signature is the commitment point, beyond which considering additional tokens hurt rather than help failure detection. In the second, persistent uncertainty, uncertainty instead accumulates throughout, and the full trace is needed to best distinguish failing from successful completions. These signatures reproduce across 23 model-dataset configurations, with the framework's falsifiable predictions holding in 20 of 23 cases, well above chance across both failure modes. Finally, we demonstrate our failure mode framework has direct implications for self-consistency, identifying when uncertainty signals complement it and when it can be selectively skipped. These results offer a foundation for understanding when LLM reasoning failures become detectable and for adapting detection strategies accordingly.
Subjects:
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Cite as: arXiv:2606.06635 [cs.CL]
(or arXiv:2606.06635v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2606.06635
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Tanvi Thoria [view email] [v1] Thu, 4 Jun 2026 18:36:42 UTC (428 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled How Language Models Fail: Token-Level Signatures of Committed and Persistent Reasoning Failures, by Tanvi Thoria and 3 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.CL
new | recent | 2026-06
Change to browse by:
cs cs.AI
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?)