Skip to main content
archive
Search Submit Donate Log in
Press Enter to search · Advanced search

Computer Science > Machine Learning

arXiv:2602.00217v1 (cs)
[Submitted on 30 Jan 2026 (this version), latest version 5 May 2026 (v3)]

Title:Dispersion Loss Counteracts Embedding Condensation and Improves Generalization in Small Language Models

Authors:Chen Liu, Xingzhi Sun, Xi Xiao, Alexandre Van Tassel, Ke Xu, Kristof Reimann, Danqi Liao, Mark Gerstein, Tianyang Wang, Xiao Wang, Smita Krishnaswamy
View a PDF of the paper titled Dispersion Loss Counteracts Embedding Condensation and Improves Generalization in Small Language Models, by Chen Liu and 10 other authors
View PDF HTML (experimental)
Abstract:Large language models (LLMs) achieve remarkable performance through ever-increasing parameter counts, but scaling incurs steep computational costs. To better understand LLM scaling, we study representational differences between LLMs and their smaller counterparts, with the goal of replicating the representational qualities of larger models in the smaller models. We observe a geometric phenomenon which we term $\textbf{embedding condensation}$, where token embeddings collapse into a narrow cone-like subspace in some language models. Through systematic analyses across multiple Transformer families, we show that small models such as $\texttt{GPT2}$ and $\texttt{Qwen3-0.6B}$ exhibit severe condensation, whereas the larger models such as $\texttt{GPT2-xl}$ and $\texttt{Qwen3-32B}$ are more resistant to this phenomenon. Additional observations show that embedding condensation is not reliably mitigated by knowledge distillation from larger models. To fight against it, we formulate a dispersion loss that explicitly encourages embedding dispersion during training. Experiments demonstrate that it mitigates condensation, recovers dispersion patterns seen in larger models, and yields performance gains across 10 benchmarks. We believe this work offers a principled path toward improving smaller Transformers without additional parameters.
Subjects: Machine Learning (cs.LG)
Cite as: arXiv:2602.00217 [cs.LG]
  (or arXiv:2602.00217v1 [cs.LG] for this version)
  https://doi.org/10.48550/arXiv.2602.00217
arXiv-issued DOI via DataCite

Submission history

From: Chen Liu [view email]
[v1] Fri, 30 Jan 2026 16:07:03 UTC (4,319 KB)
[v2] Mon, 4 May 2026 00:25:27 UTC (5,504 KB)
[v3] Tue, 5 May 2026 01:31:55 UTC (5,504 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Dispersion Loss Counteracts Embedding Condensation and Improves Generalization in Small Language Models, by Chen Liu and 10 other authors
  • View PDF
  • HTML (experimental)
  • TeX Source
view license

Current browse context:

cs.LG
< prev   |   next >
new | recent | 2026-02
Change to browse by:
cs

References & Citations

  • NASA ADS
  • Google Scholar
  • Semantic Scholar
Loading...

BibTeX formatted citation

Data provided by:

Bookmark

BibSonomy Reddit

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?)
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?)
IArxiv Recommender (What is IArxiv?)
  • 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?)
We gratefully acknowledge support from our major funders, member institutions, , and all contributors.
About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab)
Major funding support from
Simons Foundation Simons Foundation International Schmidt Sciences