Skip to content
← Back to explorer

EstLLM: Enhancing Estonian Capabilities in Multilingual LLMs via Continued Pretraining and Post-Training

Aleksei Dorkin, Taido Purason, Emil Kalbaliyev, Hele-Andra Kuulmets, Marii Ojastu, Mark Fišel, Tanel Alumäe, Eleri Aedmaa, Krister Kruusmaa, Kairit Sirts · Mar 2, 2026 · Citations: 0

Coding Math Multilingual Pairwise Preference
Open arXiv Find Implementation RSS feed Shortlist (0)

Data freshness

Extraction: Fresh

Check recency before relying on this page for active eval decisions. Use stale pages as context and verify against current hub results.

Metadata refreshed

Mar 2, 2026, 4:24 PM

Recent

Extraction refreshed

Mar 6, 2026, 11:36 PM

Fresh

Extraction source

Persisted extraction

Confidence 0.45

Abstract

Large language models (LLMs) are predominantly trained on English-centric data, resulting in uneven performance for smaller languages. We study whether continued pretraining (CPT) can substantially improve Estonian capabilities in a pretrained multilingual LLM while preserving its English and general reasoning performance. Using Llama 3.1 8B as the main base model, we perform CPT on a mixture that increases Estonian exposure while approximating the original training distribution through English replay and the inclusion of code, mathematics, and instruction-like data. We subsequently apply supervised fine-tuning, preference optimization, and chat vector merging to introduce robust instruction-following behavior. Evaluation on a comprehensive suite of Estonian benchmarks shows consistent gains in linguistic competence, knowledge, reasoning, translation quality, and instruction-following compared to the original base model and its instruction-tuned variant, while maintaining competitive performance on English benchmarks. These findings indicate that CPT, with an appropriately balanced data mixture, together with post-training alignment, can substantially improve single-language capabilities in pretrained multilingual LLMs.

Low-signal caution for protocol decisions

Use this page for context, then validate protocol choices against stronger HFEPX references before implementation decisions.

  • Extraction confidence is 0.45 (below strong-reference threshold).
  • No explicit evaluation mode was extracted from available metadata.
  • No benchmark/dataset or metric anchors were extracted.
Human Eval Hub LLM-as-Judge Hub Pairwise Preference Hub

HFEPX Relevance Assessment

This paper is adjacent to HFEPX scope and is best used for background context, not as a primary protocol reference.

Best use

Background context only

Use if you need

Background context only.

Main weakness

Extraction confidence is 0.45 (below strong-reference threshold).

Trust level

Low

Eval-Fit Score

40/100 • Low

Treat as adjacent context, not a core eval-method reference.

Human Feedback Signal

Detected

Evaluation Signal

Weak / implicit signal

HFEPX Fit

Adjacent candidate

Extraction confidence: Low

If you are doing eval pipeline work, start here:

Human Eval Hub LLM-as-Judge Hub Pairwise Preference Hub Tool-Use Eval Hub

Field Provenance & Confidence

Each key protocol field shows extraction state, confidence band, and data source so you can decide whether to trust it directly or validate from full text.

Human Feedback Types

partial

Pairwise Preference

Confidence: Low Source: Persisted extraction evidenced

Directly usable for protocol triage.

Evidence snippet: Large language models (LLMs) are predominantly trained on English-centric data, resulting in uneven performance for smaller languages.

Evaluation Modes

missing

None explicit

Confidence: Low Source: Persisted extraction missing

Validate eval design from full paper text.

Evidence snippet: Large language models (LLMs) are predominantly trained on English-centric data, resulting in uneven performance for smaller languages.

Quality Controls

missing

Not reported

Confidence: Low Source: Persisted extraction missing

No explicit QC controls found.

Evidence snippet: Large language models (LLMs) are predominantly trained on English-centric data, resulting in uneven performance for smaller languages.

Benchmarks / Datasets

missing

Not extracted

Confidence: Low Source: Persisted extraction missing

No benchmark anchors detected.

Evidence snippet: Large language models (LLMs) are predominantly trained on English-centric data, resulting in uneven performance for smaller languages.

Reported Metrics

missing

Not extracted

Confidence: Low Source: Persisted extraction missing

No metric anchors detected.

Evidence snippet: Large language models (LLMs) are predominantly trained on English-centric data, resulting in uneven performance for smaller languages.

Rater Population

missing

Unknown

Confidence: Low Source: Persisted extraction missing

Rater source not explicitly reported.

Evidence snippet: Large language models (LLMs) are predominantly trained on English-centric data, resulting in uneven performance for smaller languages.

Human Data Lens

  • Uses human feedback: Yes
  • Feedback types: Pairwise Preference
  • Rater population: Unknown
  • Unit of annotation: Unknown
  • Expertise required: Math, Coding, Multilingual
  • Extraction source: Persisted extraction

Evaluation Lens

  • Evaluation modes:
  • Agentic eval: None
  • Quality controls: Not reported
  • Confidence: 0.45
  • Flags: ambiguous, runtime_fallback_extraction

Protocol And Measurement Signals

Benchmarks / Datasets

No benchmark or dataset names were extracted from the available abstract.

Reported Metrics

No metric terms were extracted from the available abstract.

Research Brief

Deterministic synthesis

We subsequently apply supervised fine-tuning, preference optimization, and chat vector merging to introduce robust instruction-following behavior. HFEPX signals include Pairwise Preference with confidence 0.45. Updated from current HFEPX corpus.

Generated Mar 6, 2026, 11:36 PM · Grounded in abstract + metadata only

Key Takeaways

  • We subsequently apply supervised fine-tuning, preference optimization, and chat vector merging to introduce robust instruction-following behavior.
  • Evaluation on a comprehensive suite of Estonian benchmarks shows consistent gains in linguistic competence, knowledge, reasoning, translation quality, and instruction-following…

Researcher Actions

  • Compare its human-feedback setup against pairwise and rubric hubs.
  • Identify benchmark choices from full text before operationalizing conclusions.
  • Verify metric definitions before comparing against your eval pipeline.

Caveats

  • Generated from title, abstract, and extracted metadata only; full-paper implementation details are not parsed.
  • Extraction confidence is probabilistic and should be validated for critical decisions.

Research Summary

Contribution Summary

  • We subsequently apply supervised fine-tuning, preference optimization, and chat vector merging to introduce robust instruction-following behavior.
  • Evaluation on a comprehensive suite of Estonian benchmarks shows consistent gains in linguistic competence, knowledge, reasoning, translation quality, and instruction-following compared to the original base model and its instruction-tuned…

Why It Matters For Eval

  • We subsequently apply supervised fine-tuning, preference optimization, and chat vector merging to introduce robust instruction-following behavior.
  • Evaluation on a comprehensive suite of Estonian benchmarks shows consistent gains in linguistic competence, knowledge, reasoning, translation quality, and instruction-following compared to the original base model and its instruction-tuned…

Researcher Checklist

  • Pass: Human feedback protocol is explicit

    Detected: Pairwise Preference

  • Gap: Evaluation mode is explicit

    No clear evaluation mode extracted.

  • Gap: Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Gap: Benchmark or dataset anchors are present

    No benchmark/dataset anchor extracted from abstract.

  • Gap: Metric reporting is present

    No metric terms extracted.

Related Papers

Papers are ranked by protocol overlap, extraction signal alignment, and semantic proximity.

Need human evaluators for your AI research? Scale annotation with expert AI Trainers.

Post a Job Get a Quote