LmCast :: Stay tuned in

Constraint Decay: The Fragility of LLM Agents in Back End Code Generation

Recorded: May 24, 2026, 2:59 p.m.

Original Summarized

[2605.06445] Constraint Decay: The Fragility of LLM Agents in Backend Code Generation

Skip to main content

Learn about arXiv becoming an independent nonprofit.

We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors.
Donate

> cs > arXiv:2605.06445

Help | Advanced Search

All fields
Title
Author
Abstract
Comments
Journal reference
ACM classification
MSC classification
Report number
arXiv identifier
DOI
ORCID
arXiv author ID
Help pages
Full text

Search

GO

quick links

Login
Help Pages
About

Computer Science > Software Engineering

arXiv:2605.06445 (cs)

[Submitted on 7 May 2026]
Title:Constraint Decay: The Fragility of LLM Agents in Backend Code Generation
Authors:Francesco Dente, Dario Satriani, Paolo Papotti View a PDF of the paper titled Constraint Decay: The Fragility of LLM Agents in Backend Code Generation, by Francesco Dente and 2 other authors
View PDF
HTML (experimental)

Abstract:Large Language Model (LLM) agents demonstrate strong performance in autonomous code generation under loose specifications. However, production-grade software requires strict adherence to structural constraints, such as architectural patterns, databases, and object-relational mappings. Existing benchmarks often overlook these non-functional requirements, rewarding functionally correct but structurally arbitrary solutions. We present a systematic study evaluating how well agents handle structural constraints in multi-file backend generation. By fixing a unified API contract across 80 greenfield generation tasks and 20 feature-implementation tasks spanning eight web frameworks, we isolate the effect of structural complexity using a dual evaluation with end-to-end behavioral tests and static verifiers. Our findings reveal a phenomenon of constraint decay: as structural requirements accumulate, agent performance exhibits a substantial decline. Capable configurations lose 30 points on average in assertion pass rates from baseline to fully specified tasks, while some weaker configurations approach zero. Framework sensitivity analysis exposes significant performance disparities: agents succeed in minimal, explicit frameworks (e.g., Flask) but perform substantially worse on average in convention-heavy environments (e.g., FastAPI, Django). Finally, error analysis identifies data-layer defects (e.g., incorrect query composition and ORM runtime violations) as the leading root causes. This work highlights that jointly satisfying functional and structural requirements remains a key open challenge for coding agents.

Subjects:

Software Engineering (cs.SE); Artificial Intelligence (cs.AI)

Cite as:
arXiv:2605.06445 [cs.SE]

 
(or
arXiv:2605.06445v1 [cs.SE] for this version)

 
https://doi.org/10.48550/arXiv.2605.06445

Focus to learn more

arXiv-issued DOI via DataCite

Submission history From: Paolo Papotti [view email] [v1]
Thu, 7 May 2026 15:44:40 UTC (401 KB)

Full-text links:
Access Paper:

View a PDF of the paper titled Constraint Decay: The Fragility of LLM Agents in Backend Code Generation, by Francesco Dente and 2 other authorsView PDFHTML (experimental)TeX Source

view license


Current browse context:
cs.SE

< prev

  |  
next >

new
|
recent
| 2026-05

Change to browse by:

cs
cs.AI

References & Citations

NASA ADSGoogle Scholar
Semantic Scholar

export BibTeX citation
Loading...

BibTeX formatted citation
×

loading...

Data provided by:

Bookmark

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?)

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

Large Language Model agents exhibit strong proficiency in autonomous code generation when operating under loose specifications; however, the demands of production-grade software necessitate strict adherence to structural constraints, including architectural patterns, database implementations, and object-relational mappings. Existing evaluation benchmarks often fail to account for these non-functional requirements, leading to solutions that are functionally correct but structurally arbitrary. The authors conducted a systematic study to evaluate how effectively agents manage these structural constraints during multi-file backend code generation.

To isolate the effect of structural complexity, the researchers standardized the evaluation by fixing a unified API contract across eighty greenfield generation tasks and twenty feature-implementation tasks spanning eight different web frameworks. This process employed a dual evaluation methodology, utilizing both end-to-end behavioral tests and static verifiers to assess agent performance.

The study revealed a phenomenon termed constraint decay: as the structural requirements imposed on the agents accumulated, their performance experienced a substantial decline. Specifically, capable configurations demonstrated an average decrease of thirty points in assertion pass rates when moving from baseline to fully specified tasks, while some weaker configurations exhibited performance collapse, approaching zero. Further analysis of framework sensitivity demonstrated significant performance disparities, showing that agents are more successful in environments with minimal, explicit framework requirements, such as Flask, but perform substantially worse when dealing with convention-heavy environments like FastAPI or Django.

Error analysis further pointed to the root cause of failures, identifying data-layer defects, such as incorrect query composition and violations of ORM runtime rules, as the primary sources of errors. This research emphasizes that the joint satisfaction of both functional requirements and complex structural specifications remains a critical, open challenge for the development of coding agents.