arXiv — NLP / Computation & Language · · 3 min read

Where You Inject Diversity Matters: A Unified Framework for Diverse Generation

Mirrored from arXiv — NLP / Computation & Language for archival readability. Support the source by reading on the original site.

Computer Science > Computation and Language

arXiv:2606.10302 (cs)
[Submitted on 9 Jun 2026]

Title:Where You Inject Diversity Matters: A Unified Framework for Diverse Generation

View a PDF of the paper titled Where You Inject Diversity Matters: A Unified Framework for Diverse Generation, by Cheng Zhang and Rui Xin and Chudi Zhong
View PDF HTML (experimental)
Abstract:Open-ended generation tasks often require a set of meaningfully different outputs, yet large language models often produce similar generations. Existing test-time diversity methods operate at different stages of generation with varying effectiveness, but it remains unclear what design choices lead to meaningful diversity in the output. We introduce a framework that characterizes test-time diverse generation methods by the diversity source introduced during generation and provide a transmission score for measuring how effectively variation in the source reaches the final output. Guided by this framework, we propose fully automated specification-level generation methods that first generate diverse intermediate specifications and then condition on them to produce final responses. Across five open-ended tasks and four backbone models, specification-level injection improves output diversity over test-time baselines while maintaining comparable quality. Our analysis shows that successful diversity injection depends on both the diversity of the sources and their transmission to the output, highlighting source design and source-to-output realization as two key levers for building more diverse generation systems.
Subjects: Computation and Language (cs.CL)
Cite as: arXiv:2606.10302 [cs.CL]
  (or arXiv:2606.10302v1 [cs.CL] for this version)
  https://doi.org/10.48550/arXiv.2606.10302
arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Cheng Zhang [view email]
[v1] Tue, 9 Jun 2026 01:40:44 UTC (1,366 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Where You Inject Diversity Matters: A Unified Framework for Diverse Generation, by Cheng Zhang and Rui Xin and Chudi Zhong
  • View PDF
  • HTML (experimental)
  • TeX Source

Current browse context:

cs.CL
< prev   |   next >
Change to browse by:
cs

References & Citations

Loading...

BibTeX formatted citation

loading...
Data provided by:

Bookmark

BibSonomy Reddit
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?)
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.

Discussion (0)

Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.

Sign in →

No comments yet. Sign in and be the first to say something.

More from arXiv — NLP / Computation & Language