Yoruba
PulseAugur coverage of Yoruba — every cluster mentioning Yoruba across labs, papers, and developer communities, ranked by signal.
6 day(s) with sentiment data
-
AI models struggle with factual knowledge of Nigeria, new benchmark reveals
Researchers have developed NaijaFacts, a new benchmark designed to evaluate the factual knowledge of AI models about Nigeria, disentangling language proficiency from actual knowledge. The benchmark includes questions in…
-
LLM safety fails in African languages, study finds
A new study published on arXiv evaluated the safety of six large language models, including GPT-4o mini and Gemini 2.0 Flash, when tested with adversarial prompts in Yoruba and Nigerian Pidgin. The research found that E…
-
Nigerian startup releases StewMind 1.7B, a compact, local AI agent
StewMind 1.7B, a compact agentic AI model developed in Nigeria, has been released. Unlike large, cloud-based models, StewMind is designed to be downloaded and run locally on user devices, including phones, eliminating A…
-
BlockVectra MCP plugin integrates with Antigravity CLI for documentation access
A new plugin called BlockVectra MCP has been released for the Antigravity CLI, allowing users to integrate with the BlockVectra documentation server. The installation process involves adding the plugin from its GitHub r…
-
AI refusal directions show limited universality in low-resourced languages, llama.cpp updates released
A new research paper investigates the universality of AI refusal directions across low-resourced languages, extending previous work that suggested near-perfect cross-lingual transfer. The study found that while directio…
-
Yorùbá language faces digital encoding issues due to Unicode limitations
A research paper highlights a persistent issue with writing the Yorùbá language digitally due to limitations in Unicode encoding. The paper details how the lack of specific precomposed characters forces the use of incon…
-
Yorùbá language paper proposes new typographic conventions for contour tones
A new paper addresses the orthographic challenges of representing contour tones in the Yorùbá language, which can lead to misinterpretations or inversions of meaning, particularly in names. The authors propose using car…
-
Nigerian low-resource languages surveyed for NLP advancements
A new survey paper titled "NaijaNLP: A Survey of Nigerian Low-Resource Languages" has been published on arXiv, detailing the current state of natural language processing (NLP) research for three major Nigerian languages…
-
Study finds LLM data audits don't guarantee downstream utility for African NLP
A new study published on arXiv investigates the effectiveness of synthetic data selection methods for low-resource African languages. The research found that common proxies, which assume that data rated highly by an LLM…
-
New research tackles multilingual AI efficiency and capabilities
Researchers are developing new methods to improve the efficiency and capabilities of multilingual AI models. One study explores token merging for multilingual speech recognition, showing it can significantly reduce comp…
-
New evaluation method reveals ASR/audio LM struggles with English-Yoruba code-switching
A new research paper introduces a switch-aware evaluation method for automatic speech recognition (ASR) and audio language models (audio LMs) when processing code-switched speech, specifically focusing on English and Yo…
-
First City Monument Bank to offer AI advisory in local languages for farmers
First City Monument Bank is integrating AI to offer advisory services in Hausa, Yoruba, and Igbo languages. This initiative aims to support smallholder farmers. The bank's efforts are detailed in its Agritech Alumni Imp…
-
New method recovers AI safety for African languages without retraining
Researchers have developed a novel training-free method called Latent Space Refusal Anchoring (LSR-Anchoring) to improve safety in instruction-tuned AI models for low-resource African languages. This technique aims to r…
-
New AfriNLLB models offer efficient translation for 15 African languages
Researchers have developed AfriNLLB, a suite of lightweight translation models designed for African languages. These models are derived from the NLLB-200 600M architecture, which has been compressed through layer prunin…
-
Tokenization premiums create AI cost barriers for non-English languages · arXiv cs.CL
A new study published on arXiv introduces the Tokenization Equity Audit (TEA), a benchmark designed to measure disparities in how large language models tokenize different languages. The research found that semantically …
-
AI's impact on linguistic diversity in World Englishes debated in new papers · 3 sources tracked
Three recent academic papers explore the complex relationship between generative AI and linguistic diversity, particularly concerning World Englishes. The first paper discusses how AI tools can both democratize academic…
-
New speech synthesizer developed for Yoruba language
Researchers have developed TTSYoruba, a rule-based speech synthesizer for the Yoruba language. This system utilizes a hand-crafted phonological rule system and a recorded inventory of diphone units to convert tone-marke…
-
Developer builds AI with memory system, using specialized Qwen models
A developer has created a novel AI memory system named Àtúnbí, meaning "reborn" in Yoruba, designed to overcome the limitations of stateless AI conversations. This system prioritizes information based on importance and …
-
OmniVoice fine-tuned for Yoruba zero-shot voice cloning
A developer fine-tuned the OmniVoice text-to-speech model for the Yoruba language, a tonal language where precise pronunciation is critical for meaning. The process involved constructing a dataset by merging high-qualit…
-
NSFW Anime Illustrations Promoted on Mastodon
This cluster contains two Mastodon posts featuring anime-style illustrations of characters, Rikka Takarada and Yor. The posts include hashtags related to NSFW content, hentai, and AI-generated imagery. They also promote…