Skip to content
OpenTrain AIFor AI Companies

HFEPX · Eval paper review

Show Your Work in Swahili: Reinforcement Learning for Native-Language Mathematical Reasoning

Jessica Lundin, Katherine Rosenfeld, David Adelani, Geoffrey Mosoti Nyakiongora +4 more

Published

Nov 11, 2026

Citations

0

Trust level

Moderate

Usefulness score

37/100 (Low)

Extraction confidence

55% (Moderate)

Derived from extracted protocol signals and abstract evidence.

Rater population

Not reported

Signals refreshed

Nov 11, 2026

Open arXiv Find implementation Shortlist (0) RSS feed

Tags

Automatic Metrics Human Eval Math

Should you rely on this paper?

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

Use this for comparison and orientation, not as your only source.

Best use

Background context only

Use if you need

A benchmark-and-metrics comparison anchor.

What to verify

Validate the evaluation procedure and quality controls in the full paper before operational use.

Main weakness

No major weakness surfaced.

Human feedback signal
Not explicit
Not explicit in abstract metadata
Evaluation signal
Detected
Eval setup described
Usefulness for eval research
37/100
Adjacent candidate

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

If you are doing eval pipeline work, start here

Human Eval Papers LLM-as-Judge Papers Pairwise Preference Papers Tool-Use Evaluation Search

Abstract

We present Fikiria, a two-stage SFT+GRPO recipe for Swahili mathematical reasoning. Its reward multiplies a format gate by a weighted sum of answer accuracy and the ChrF\texttt{++} similarity of the Swahili reasoning to an in-language reference chain of thought, so that training targets the surface fidelity of the reasoning trace and not only the final answer. We train on a 2.3k-problem Swahili dataset created by translating GSM8K with GPT-4.1 and filtering with SSA-COMET. On AfriMGSM Swahili, Fikiria-AfriqueQwen-14B reaches 79.2\%. Under the same fixed prompt and strict answer-format scoring, GPT-4o reaches 68.0\%, partly because it often fails the required answer format; IrokoBench reports 78.8\% for GPT-4o using a best-per-model prompt, a figure numerically close to ours but not directly comparable. In a small human evaluation with co-author annotators, GPT-4o's outputs were preferred overall. ChrF\texttt{++} measures surface fidelity, not the validity of individual reasoning steps.

What we could verify

These are the protocol signals we could actually recover from the available paper metadata. Use them to decide whether this paper is worth deeper reading.

Human Feedback Types

missing

None explicit

No explicit feedback protocol extracted.

"We present Fikiria, a two-stage SFT+GRPO recipe for Swahili mathematical reasoning."

Evaluation Modes

strong

Human Eval, Automatic Metrics

Includes extracted eval setup.

"We present Fikiria, a two-stage SFT+GRPO recipe for Swahili mathematical reasoning."

Quality Controls

missing

Not reported

No explicit QC controls found.

"We present Fikiria, a two-stage SFT+GRPO recipe for Swahili mathematical reasoning."

Benchmarks / Datasets

strong

GSM8K

Useful for quick benchmark comparison.

"We train on a 2.3k-problem Swahili dataset created by translating GSM8K with GPT-4.1 and filtering with SSA-COMET."

Reported Metrics

strong

Accuracy

Useful for evaluation criteria comparison.

"Its reward multiplies a format gate by a weighted sum of answer accuracy and the ChrF\texttt{++} similarity of the Swahili reasoning to an in-language reference chain of thought, so that training targets the surface fidelity of the reasoning trace and not only the final answer."

Benchmarks and datasets

GSM8K

Reported metrics

accuracy
Human feedback details
Uses human feedback
No
Feedback types
None
Rater population
Not reported
Expertise required
Math
Evaluation details
Evaluation modes
Human Eval, Automatic Metrics
Agentic eval
None
Quality controls
Not reported
Evidence quality
Moderate
Use this page as
Background context only

Research brief

Metadata summary

We present Fikiria, a two-stage SFT+GRPO recipe for Swahili mathematical reasoning.

Based on abstract + metadata only. Check the source paper before making high-confidence protocol decisions.

Key takeaways

  • We present Fikiria, a two-stage SFT+GRPO recipe for Swahili mathematical reasoning.
  • Its reward multiplies a format gate by a weighted sum of answer accuracy and the ChrF\texttt{++} similarity of the Swahili reasoning to an in-language reference chain of thought, so that training targets the surface fidelity of the reasoning trace and not only the final answer.
  • We train on a 2.3k-problem Swahili dataset created by translating GSM8K with GPT-4.1 and filtering with SSA-COMET.

Researcher actions

  • Compare this paper against others mentioning GSM8K.
  • Validate inferred eval signals (Human evaluation, Automatic metrics) against the full paper.
  • Use related-paper links to find stronger protocol-specific references.

Caveats

  • Generated from abstract + metadata only; no PDF parsing.
  • Signals below are heuristic and may miss details reported outside the abstract.

Recommended queries

GSM8K Accuracy

Contribution summary

  • We present Fikiria, a two-stage SFT+GRPO recipe for Swahili mathematical reasoning.
  • Its reward multiplies a format gate by a weighted sum of answer accuracy and the ChrF\texttt{++} similarity of the Swahili reasoning to an in-language reference chain of thought, so that training targets the surface fidelity of the…

Why it matters for eval

  • In a small human evaluation with co-author annotators, GPT-4o's outputs were preferred overall.

Researcher checklist

  • Human feedback protocol is explicit

    No explicit human feedback protocol detected.

  • Evaluation mode is explicit

    Detected: Human Eval, Automatic Metrics

  • Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Benchmark or dataset anchors are present

    Detected: GSM8K

  • Metric reporting is present

    Detected: accuracy