Open to Work

5 45 97

Mahamadi NIKIEMA

madoss

https://mnikiema.github.io/

AI & ML interests

AI & ML Engineer focused on low-resource languages. Building multilingual NLP, ASR, and TTS systems for Mooré and French. Fine-tuning LLMs and speech models, curating parallel corpora, and shipping reproducible pipelines. Open to work.

Recent Activity

liked a dataset 5 days ago

espnet/yodas-granary

liked a model 5 days ago

Archime/parakeet-tdt-0.6b-v3-fr-tv-media

updated a collection 6 days ago

SLM

View all activity

Organizations

upvoted a paper 16 days ago

LFM2 Technical Report

Paper • 2511.23404 • Published Nov 28, 2025 • 59

upvoted a collection 19 days ago

BidirLM

Collection

BidirLM is a family of 5 frontier bidirectional encoders, including an omnimodal variant at 2.5B. • 8 items • Updated 30 days ago • 1

upvoted an article 24 days ago

Article

DenseOn with the LateOn: Open State-of-the-Art Single and Multi-Vector Models

lightonai

•

24 days ago

• 38

upvoted an article 26 days ago

Article

Building a Fast Multilingual OCR Model with Synthetic Data

nvidia

•

28 days ago

• 33

upvoted an article 28 days ago

Article

Fine-Tune W2V2-Bert for low-resource ASR with 🤗 Transformers

ylacombe

•

Jan 19, 2024

• 48

upvoted 2 papers 28 days ago

Embarrassingly Simple Self-Distillation Improves Code Generation

Paper • 2604.01193 • Published Apr 1 • 50

TRACER: Trace-Based Adaptive Cost-Efficient Routing for LLM Classification

Paper • 2604.14531 • Published 29 days ago • 7

upvoted a paper 29 days ago

Adapting LLaMA Decoder to Vision Transformer

Paper • 2404.06773 • Published Apr 10, 2024 • 19

upvoted an article about 1 month ago

Article

Mastering Tensor Dimensions in Transformers

not-lain

•

Jan 12, 2025

• 172

upvoted an article about 2 months ago

Article

How I contributed a new model to the Transformers library using Codex

nielsr

•

Mar 30

• 51

upvoted a collection about 2 months ago

fiNERweb

Collection

A multilingual dataset for NER covering 91 langauges and 25 scripts • 3 items • Updated Dec 16, 2025 • 3

upvoted an article about 2 months ago

Article

Introducing AI chunking to semchunk

isaacus

•

Mar 23

• 9

upvoted a paper about 2 months ago

Omnilingual MT: Machine Translation for 1,600 Languages

Paper • 2603.16309 • Published Mar 17 • 22

upvoted a collection 2 months ago

Fine-tune ready versions of the LLMSQL benchmark

Collection

This collection contains the versions of the benchmark in fine-tune ready format • 2 items • Updated Mar 4 • 1

upvoted a paper 2 months ago

Beyond Language Modeling: An Exploration of Multimodal Pretraining

Paper • 2603.03276 • Published Mar 3 • 104

upvoted an article 2 months ago

Article

DABStep: Data Agent Benchmark for Multi-step Reasoning

eggie5, martinigoyanes, frisokingma, andreumora, lvwerra, thomwolf, m-ric

•

Feb 4, 2025

• 130

upvoted an article 3 months ago

Article

Mixture of Experts (MoEs) in Transformers

ariG23498, pcuenq, merve, IlyasMoutawwakil, ArthurZ, sergiopaniego, Molbap

•

Feb 26

• 159

upvoted 3 papers 3 months ago

The Million-Label NER: Breaking Scale Barriers with GLiNER bi-encoder

Paper • 2602.18487 • Published Feb 11 • 6

Optimal Turkish Subword Strategies at Scale: Systematic Evaluation of Data, Vocabulary, Morphology Interplay

Paper • 2602.06942 • Published Feb 6 • 3

Qwen3-ASR Technical Report

Paper • 2601.21337 • Published Jan 29 • 37

Mahamadi NIKIEMA

AI & ML interests

Recent Activity

Organizations

madoss's activity

DenseOn with the LateOn: Open State-of-the-Art Single and Multi-Vector Models

Building a Fast Multilingual OCR Model with Synthetic Data

Fine-Tune W2V2-Bert for low-resource ASR with 🤗 Transformers

Mastering Tensor Dimensions in Transformers

How I contributed a new model to the Transformers library using Codex

Introducing AI chunking to semchunk

DABStep: Data Agent Benchmark for Multi-step Reasoning

Mixture of Experts (MoEs) in Transformers