Turkish
PulseAugur coverage of Turkish — every cluster mentioning Turkish across labs, papers, and developer communities, ranked by signal.
5 day(s) with sentiment data
-
AI agent Hermes connects to agenzax, researches global suppliers in native languages
The author describes their experience connecting their AI agent, Hermes, to a new platform called agenzax, which allows agents to interact with the real world and each other. Initially hesitant to let their agent operat…
-
Psychosis linked to information compression deficit in speech, study finds
A new study published on arXiv explores the link between psychosis and information compression in language. Researchers analyzed speech from Turkish speakers, including those with schizophrenia spectrum disorders, and f…
-
LLMs struggle with inferential narrative features in Turkish corpus study
A new study evaluated the inter-rater reliability of large language models (LLMs) and rule-based systems in annotating inferential narrative features within a Turkish corpus. The research found that models like Gemini 2…
-
Developer creates Kip and Tenioha languages using grammatical cases and particles
A developer created a functional programming language named Kip, which incorporates Turkish grammatical cases into its type system. Inspired by this, the developer then built a similar language called Tenioha, designed …
-
New methods improve multilingual video transcription accuracy
Researchers have developed methods to improve speech transcription accuracy from videos across multiple languages, aiming to aid the creation of automated tools for cross-cultural understanding. By leveraging publicly a…
-
ChatGPT recommendations influenced by query language, not location
A study investigating ChatGPT's commercial recommendations found that query language, rather than exit IP location, primarily determines whether local suppliers are featured. The research, conducted across multiple lang…
-
MoganColBERT-TR: New Turkish Multi-Vector Retrieval Model Unveiled
Researchers have developed MoganColBERT-TR, a new multi-vector retrieval model specifically designed for the Turkish language. This model builds upon a previously trained ModernBERT encoder and adapts it to the ColBERT …
-
New Turkish LLM MoganBert-TR trained with CLM-to-MLM curriculum
Researchers have developed MoganBert-TR, a new Turkish encoder foundation model, and its accompanying embedding model, MoganBert-Embed. Trained from scratch on a filtered Turkish corpus using a novel CLM-to-MLM curricul…
-
New RAG evaluation methods emerge for Turkish and domain-specific data · 4 sources tracked
Researchers are developing new methods to evaluate and improve Retrieval-Augmented Generation (RAG) systems. One study compares different chunking and embedding strategies for Turkish RAG, finding that layout-aware chun…
-
New Turkish multimodal corpus aims to improve conversational AI turn-taking
Researchers have introduced Real-TurnTurk, a new multimodal Turkish conversational dataset designed to improve turn-taking prediction in synchronous dialogue systems. The dataset comprises synchronized video, audio, and…
-
Prompt engineering: Write output-like text natively, instructions can be translated
A new approach to prompt engineering suggests that only the parts of a prompt resembling the desired output should be written natively in the target language, while instructions and machinery can remain in a language th…
-
New method learns sign language representations from broadcast news
Researchers have developed a method for learning sign language representations from broadcast news transcripts, which offer weak supervision due to loose alignment between spoken words and signing. This approach is part…
-
GFlowNets used to generate novel LLM attacks in English and Turkish
Researchers have developed a novel method using GFlowNets to automatically generate adversarial attacks against Large Language Models (LLMs). This approach trains an attacker model to identify vulnerabilities in a victi…
-
AssemblyAI launches Universal-3.5 Pro with expanded multilingual transcription
AssemblyAI has released its Universal-3.5 Pro model, enhancing its multilingual transcription capabilities. This new model supports automatic language detection and native code-switching across 18 languages, a significa…
-
AI agent's prompt injection detector fails on non-English attacks
A security audit of an open-source agent framework revealed a significant vulnerability in its prompt injection detection system. The scanner, which inspects context files, memory writes, and tool outputs, failed to det…
-
HukukBERT: New Language Model for Turkish Legal Texts
Researchers have developed HukukBERT, a specialized language model designed to process Turkish legal texts. This model was trained on a substantial corpus of Turkish legal documents using a hybrid pre-training approach …
-
Cross-lingual transfer in Turkic languages shows strong pair-specific performance
Researchers have investigated cross-lingual transfer techniques for machine translation within the Turkic language family, focusing on Turkish, Azerbaijani, Uzbek, Kazakh, and Kyrgyz. Their findings indicate that transf…
-
Research paper flags structural flaws in LLM-as-Judge synthetic corpora
A new research paper highlights a critical issue in creating synthetic datasets for evaluating Large Language Models (LLMs) used as judges. The study reveals that the process of generating 'hallucinated' answers for the…
-
Healthcare LLMs show significant cross-lingual factual disparities, paper finds
A new arXiv paper highlights significant disparities in the factual accuracy of Large Language Models (LLMs) when answering healthcare-related questions across different languages. Researchers developed a multilingual d…
-
LLMs' grasp of sarcasm questioned in Turkish language study
Large Language Models (LLMs) may struggle with understanding sarcasm due to its reliance on context and linguistic nuances, unlike simpler tasks like sentiment analysis. A project at Acıbadem University, in collaboratio…