arXiv cs.CL· Tohir Saidzoda, Nekruz Maxudov, Bedil Karimov, Abubakr Abdullozoda, Amin Jalilov, Jonibek Kishvarov, Rahmonali Sharipov, Mariyamkhon Shodyeva·· 11 小时前AI 评分31
Saidzoda-Tajik-LFM-2.6B：将 2.6B 混合模型适配塔吉克语工具调用，以及持续预训练会带走什么
Saidzoda-Tajik-LFM-2.6B: Adapting a 2.6B Hybrid Model to Tajik Tool Calling, and What Continued Pretraining Takes Away
AI 导读
研究者发布 Saidzoda-Tajik-LFM-2.6B，将 2.6B 参数的混合模型 LFM2.5-2.6B-Base 适配塔吉克语工具调用：先对 1,485,319 篇塔吉克语网页文档做 3 个 epoch 的 LoRA 持续预训练，再做 1 个 epoch SFT。
正文 · 原文
Abstract:We release Saidzoda-Tajik-LFM-2.6B, a Tajik adaptation of the 2.6B-parameter hybrid model LFM2.5-2.6B-Base, built for tool calling in Tajik on modest hardware. The recipe has two stages: LoRA continued pretraining (CPT) on 1,485,319 cleaned Tajik web documents for 3 epochs, then one epoch of supervised fine-tuning (SFT) on a mix of machine-translated Tajik tool calls, English tool calls kept as replay, and Tajik instructions. On a 174-item tool-calling test set built from BFCL v3 and machine-translated to Tajik, strict full-call accuracy on Tajik prompts rises from 20.69 for the base model to 64.14 (paired difference +43.45 points, 95% interval +35.2 to +51.7); a lenient parser that also accepts the base model's own call formats puts the base at 31.72. English shows no detectable change (76.55 and 77.24). Every CPT checkpoint scores zero: CPT removes the native tool-call format and, as the lenient re-scoring shows, part of the skill itself. Outside tool calling, accuracy on a Tajik knowledge benchmark rises from 28.0 to 35.0, Belebele Tajik shows no significant change, and English MMLU falls from 60.8 to 52.5. The model is not a general Tajik assistant. It almost never declines when no tool fits (6.9% correct on Tajik, against 31.0% for the base under the lenient parser), it is weak at multi-turn calls, and its calls should be checked before they are executed. We describe the data, the training settings and every evaluation in enough detail to repeat them.
| Comments: | 12 pages, 9 tables, 3 figures. Technical report. Model: this https URL |
| Subjects: | Computation and Language (cs.CL) |
| ACM classes: | I.2.7 |
| Cite as: | arXiv:2610.41576 [cs.CL] |
| (or arXiv:2610.41576v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.41576 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Tohir Saidzoda Mr [view email]
[v1]
Fri, 30 Oct 2026 11:06:30 UTC (134 KB)
来源：arXiv cs.CL · arxiv.org