Skip to content
AI News HubLIVE
Original source2 min read

Selecting Open-Weight Language Models for Zero-Shot Intent Classification: A Systematic Evaluation of 41 Models

Summary

An arXiv paper systematically evaluates 41 open-weight language models for zero-shot intent classification across 15 families and 135M–9B parameters on eight English single-label datasets, plus an auxiliary five-shot ATIS result. Beyond accuracy, it analyzes confidence calibration, perturbation robustness, statistical reliability of rankings, deployment efficiency, and benchmark saturation, finding that instruction-tuned 3B models can beat several 7B base models, that top models on MASSIVE are statistically indistinguishable under McNemar tests, and that benchmarks like SNIPS are saturated.

SourcearXiv Computational LinguisticsAuthor: Parishruthi Ganesh, Gerry Dozier, Cheryl Seals
Selecting Open-Weight Language Models for Zero-Shot Intent Classification: A Systematic Evaluation of 41 Models
Report an error

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

Correction instructions
Read article

[Submitted on 29 Jul 2026]

Title:Selecting Open-Weight Language Models for Zero-Shot Intent Classification: A Systematic Evaluation of 41 Models

View a PDF of the paper titled Selecting Open-Weight Language Models for Zero-Shot Intent Classification: A Systematic Evaluation of 41 Models, by Parishruthi Ganesh and 2 other authors

View PDF HTML (experimental)

Abstract:Intent classification is a core component of task-oriented dialogue systems, yet practitioners have limited systematic guidance for selecting deployable open-weight language models under compute, latency, and robustness constraints. We present a systematic zero-shot evaluation of 41 open-weight language models spanning 15 families and the 135M--9B parameter range across eight English single-label intent-classification datasets. A ninth dataset, ATIS, uses five labeled demonstrations and is reported as an auxiliary five-shot result. The evaluation includes standard benchmarks, a large-scale voice-assistant corpus, and production-derived e-commerce datasets. Beyond exact-match accuracy, we analyze confidence calibration, robustness to realistic input perturbations, statistical reliability of model rankings, deployment efficiency, and benchmark saturation. Our results show that instruction-tuned 3B models can outperform several evaluated 7B base models, that differences among leading models on MASSIVE are statistically indistinguishable under pairwise McNemar tests, and that widely used benchmarks such as SNIPS have become saturated and no longer meaningfully discriminate among current open-weight models. Instruction tuning's effect on confidence calibration is inconsistent rather than uniformly harmful. These findings provide practical guidance for selecting and evaluating open-weight language models for intent classification.

Subjects:

Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)

Cite as: arXiv:2607.27421 [cs.CL]

(or arXiv:2607.27421v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2607.27421

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Parishruthi Ganesh [view email] [v1] Wed, 29 Jul 2026 19:42:33 UTC (898 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Selecting Open-Weight Language Models for Zero-Shot Intent Classification: A Systematic Evaluation of 41 Models, by Parishruthi Ganesh and 2 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CL

new | recent | 2026-07

Change to browse by:

cs cs.AI cs.LG

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

InvestorsAdvanced

Key points

  • Evaluates 41 open-weight models from 15 families spanning 135M–9B parameters on standard benchmarks, a large voice-assistant corpus, and production e-commerce datasets.
  • Instruction-tuned 3B models can outperform several evaluated 7B base models in zero-shot intent classification.
  • Differences among leading models on MASSIVE are statistically indistinguishable under pairwise McNemar tests, and SNIPS and similar benchmarks are saturated.
  • Instruction tuning's effect on confidence calibration is inconsistent rather than uniformly harmful, so selection should weigh accuracy, calibration, robustness, and deployment efficiency together.

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