Hugging Face Daily Papers · · 5 min read

No Resource, No Benchmarks, No Problem? Evaluating and Improving LLMs for Code Generation in No-Resource Languages

Mirrored from Hugging Face Daily Papers for archival readability. Support the source by reading on the original site.

.</p>\n","updatedAt":"2026-06-19T09:03:34.024Z","author":{"_id":"663486a1f64712540644cb68","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/663486a1f64712540644cb68/YZFR41ERY6UrC6rCC6Nan.jpeg","fullname":"Alessandro","name":"Devy1","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":4,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"fr","probability":0.32275810837745667},"editors":["Devy1"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/663486a1f64712540644cb68/YZFR41ERY6UrC6rCC6Nan.jpeg"],"reactions":[],"isReport":false}},{"id":"6a3584728567badf759a0c86","author":{"_id":"6960eca92f7ad9b043b5cbe0","avatarUrl":"/avatars/e68dcc7fd04f143d849d40414866e633.svg","fullname":"Noah","name":"noahml","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":0,"isUserFollowing":false},"createdAt":"2026-06-19T18:03:30.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"Cool paper - I liked the way \"No Resource, No Benchmarks, No Problem? Evaluating and Improving LLMs for Code Generation in No-Resource Languages\" frames the problem without making it feel too abstract.\n\nCurious if you think this would still work once the setup gets messier in the wild?\n\nI made a podcast on it with ResearchPod, it makes it easy to get the key concepts on the go:\nhttps://researchpod.app/episode/135c71ee-125b-43bd-b92c-bebd114dfdbe","html":"<p>Cool paper - I liked the way \"No Resource, No Benchmarks, No Problem? Evaluating and Improving LLMs for Code Generation in No-Resource Languages\" frames the problem without making it feel too abstract.</p>\n<p>Curious if you think this would still work once the setup gets messier in the wild?</p>\n<p>I made a podcast on it with ResearchPod, it makes it easy to get the key concepts on the go:<br><a href=\"https://researchpod.app/episode/135c71ee-125b-43bd-b92c-bebd114dfdbe\" rel=\"nofollow\">https://researchpod.app/episode/135c71ee-125b-43bd-b92c-bebd114dfdbe</a></p>\n","updatedAt":"2026-06-19T18:03:30.120Z","author":{"_id":"6960eca92f7ad9b043b5cbe0","avatarUrl":"/avatars/e68dcc7fd04f143d849d40414866e633.svg","fullname":"Noah","name":"noahml","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":0,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8500277400016785},"editors":["noahml"],"editorAvatarUrls":["/avatars/e68dcc7fd04f143d849d40414866e633.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.16827","authors":[{"_id":"6a33930559127a45e2c1c6dc","user":{"_id":"663486a1f64712540644cb68","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/663486a1f64712540644cb68/YZFR41ERY6UrC6rCC6Nan.jpeg","isPro":false,"fullname":"Alessandro","user":"Devy1","type":"user","name":"Devy1"},"name":"Alessandro Giagnorio","status":"claimed_verified","statusLastChangedAt":"2026-06-18T11:26:26.695Z","hidden":false},{"_id":"6a33930559127a45e2c1c6dd","name":"Alberto Martin-Lopez","hidden":false},{"_id":"6a33930559127a45e2c1c6de","name":"Gabriele Bavota","hidden":false}],"publishedAt":"2026-06-15T00:00:00.000Z","submittedOnDailyAt":"2026-06-19T00:00:00.000Z","title":"No Resource, No Benchmarks, No Problem? Evaluating and Improving LLMs for Code Generation in No-Resource Languages","submittedOnDailyBy":{"_id":"663486a1f64712540644cb68","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/663486a1f64712540644cb68/YZFR41ERY6UrC6rCC6Nan.jpeg","isPro":false,"fullname":"Alessandro","user":"Devy1","type":"user","name":"Devy1"},"summary":"Large Language Models (LLMs) have significantly advanced the automation of software engineering tasks. One prominent example is code generation, where an LLM produces code in a specified programming language based on a natural language description. Most research in this area has focused on high-resource languages, such as Python or Java, which benefit from abundant training data. A smaller body of work has explored low-resource languages, which are underrepresented in training corpora. In contrast, no-resource languages for which LLMs have seen virtually no training data remain largely unstudied. These languages often emerge in industry, where organizations develop proprietary or domain-specific languages unsupported by commercial tools like GitHub Copilot. This results in the need for companies to deploy their own in-house code recommenders. To investigate possible solutions in this context, we build and release three code generation benchmarks for no-resource languages, based on two recently proposed programming languages for which very little training data is available. Using these benchmarks, we experiment several solutions to teach LLMs about no-resource languages, including prompt-based techniques as well as pre-training and fine-tuning exploiting the little data available. While further pre-training gives the largest performance gains for no-resource languages, applying it directly to instruction-tuned models harms their ability to follow instructions. To address this, we start from a base model, further pre-training it on the target language, and then inject instruction-following capabilities via weight diff transfer from an instruction model. Such an approach significantly improves code generation capabilities in no-resource settings, allowing companies to cheaply deploy a specialized instruct model without dealing with the computational cost of instruction fine-tuning.","upvotes":0,"discussionId":"6a33930559127a45e2c1c6df","githubRepo":"https://github.com/Devy99/no-resource-pl-study","githubRepoAddedBy":"user","ai_summary":"Research addresses code generation challenges for no-resource programming languages by developing benchmarks and proposing a method that combines further pre-training with weight difference transfer to create specialized instruction-following models at reduced computational cost.","ai_keywords":["large language models","code generation","no-resource languages","prompt-based techniques","pre-training","fine-tuning","instruction-tuned models","weight diff transfer","instruction-following capabilities"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":1},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[],"acceptLanguages":["en"],"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.16827.md","query":{}}">
Papers
arxiv:2606.16827

No Resource, No Benchmarks, No Problem? Evaluating and Improving LLMs for Code Generation in No-Resource Languages

Published on Jun 15
· Submitted by
Alessandro
on Jun 19
Authors:
,

Abstract

Research addresses code generation challenges for no-resource programming languages by developing benchmarks and proposing a method that combines further pre-training with weight difference transfer to create specialized instruction-following models at reduced computational cost.

Large Language Models (LLMs) have significantly advanced the automation of software engineering tasks. One prominent example is code generation, where an LLM produces code in a specified programming language based on a natural language description. Most research in this area has focused on high-resource languages, such as Python or Java, which benefit from abundant training data. A smaller body of work has explored low-resource languages, which are underrepresented in training corpora. In contrast, no-resource languages for which LLMs have seen virtually no training data remain largely unstudied. These languages often emerge in industry, where organizations develop proprietary or domain-specific languages unsupported by commercial tools like GitHub Copilot. This results in the need for companies to deploy their own in-house code recommenders. To investigate possible solutions in this context, we build and release three code generation benchmarks for no-resource languages, based on two recently proposed programming languages for which very little training data is available. Using these benchmarks, we experiment several solutions to teach LLMs about no-resource languages, including prompt-based techniques as well as pre-training and fine-tuning exploiting the little data available. While further pre-training gives the largest performance gains for no-resource languages, applying it directly to instruction-tuned models harms their ability to follow instructions. To address this, we start from a base model, further pre-training it on the target language, and then inject instruction-following capabilities via weight diff transfer from an instruction model. Such an approach significantly improves code generation capabilities in no-resource settings, allowing companies to cheaply deploy a specialized instruct model without dealing with the computational cost of instruction fine-tuning.

Community

Paper author Paper submitter about 13 hours ago

.

Cool paper - I liked the way "No Resource, No Benchmarks, No Problem? Evaluating and Improving LLMs for Code Generation in No-Resource Languages" frames the problem without making it feel too abstract.

Curious if you think this would still work once the setup gets messier in the wild?

I made a podcast on it with ResearchPod, it makes it easy to get the key concepts on the go:
https://researchpod.app/episode/135c71ee-125b-43bd-b92c-bebd114dfdbe

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images

· Sign up or log in to comment

Get this paper in your agent:

hf papers read 2606.16827
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2606.16827 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2606.16827 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2606.16827 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

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 Hugging Face Daily Papers