Natural Language Processing

A new study from arXiv (2608.18142v1) finds that Parameter-Efficient Fine-Tuning (PEFT) with Low-Rank Adaptation (LoRA) significantly improves hate speech detection in Roman Urdu, boosting F1 scores from 0.56 in zero-sho…

A new method called Role-Conditioned Sub-Token Routing (RoleSub) compresses vision-language-action (VLA) model representations, outperforming token-only pruning in 33 of 36 settings on the LIBERO suites when tested with …

A scoping review published on arXiv (2608.18135v1) found that artificial intelligence technologies integrated into rural medication management reduce prescribing and transcription errors by 34% to 80%, while facing barri…

Researchers introduced StocksTalk, a voice-enabled conversational system that converts spoken financial screening requests into executable SQL queries over real-world market data, combining streaming speech recognition, …

A new arXiv study (2608.18085v1) finds that large language models can express personality traits in task-oriented dialogue while maintaining strong task performance, but adapting to the user's personality improves constr…

Researchers introduced BERTilda, an explainable framework that tracks topic lifecycles in longitudinal text streams by detecting splits, merges, continuations, and disappearances using similarity-and-flow temporal graphs…

Researchers introduced NE-BERT, a multilingual encoder model trained on 8.3 million sentences across 9 Northeast Indian languages plus Hindi and English, which outperforms IndicBERT-V2 and MuRIL on all 9 languages with 1…

Researchers introduced FrenchNews-7, a cross-publisher benchmark for classifying French-language news articles into seven editorial desk categories, featuring a fine-tuned CamemBERT classifier that achieved an overall re…

A new arXiv study (2608.18158v1) evaluating LLMs as data quality annotators found that on the Abt Buy entity matching benchmark (2,194 labeled pairs), a rule-based baseline (F1=0.950) performed about as well as LLM zero-…

GreekBarRetrieval, a new public benchmark for Greek statutory retrieval derived from GreekBarBench, comprises 283 bar-exam questions and 6,308 candidate statutory articles. Testing three BM25 variants and nine dense retr…

Klo, a native macOS video editor driven by an AI agent, launched on Hacker News after 40 releases in 30 days, allowing users to edit videos via plain English commands such as 'cut the dead air' or 'caption it word by wor…

A developer built an AI Pharmacist Assistant that uses OCR to scan drug labels and retrieval-augmented generation (RAG) to detect drug-drug interactions. The system extracts drug names from images, cross-references a SQL…

Manticore Search introduced Conversational Search, a feature that combines full-text, vector, and hybrid search with a language model to answer complex product queries, as demonstrated in its Manticore Apparel Shop demo …

A feature built by Soamee for a SaaS client, costing less than $30 per month to run, reduced support tickets by 40% in the first eight weeks. The system uses basic RAG with OpenAI's text-embedding-3-small and Claude Haik…

Kristyna Onderkova, a researcher presenting at the 1st Workshop for Young Researchers in Natural Language Generation (YNLG) in Hanoi, Vietnam, in October 2025, described her work on improving factual accuracy and depth i…

Nalin Kumar's position paper, presented at the 1st Workshop for Young Researchers in Natural Language Generation (YNLG) in Hanoi, Vietnam, in October 2025, proposes two efficient strategies for building large language mo…

Barkavi Sundararajan's PhD research, presented at the 1st Workshop for Young Researchers in Natural Language Generation in Hanoi, Vietnam, in October 2025, proposes two approaches to reduce hallucinations in neural data-…

Researchers from the University of Amsterdam and other institutions propose LINK, a data-level intervention method that improves cross-lingual knowledge transfer in multilingual language models by substituting random wor…

Healthwatch Rotherham reported that patients in Yorkshire are struggling to use an AI GP receptionist called EMMA because it cannot understand broad Yorkshire accents, causing frustration and leading some to visit surger…

Google has integrated Gemini Live with its Deep Research capability, enabling users to initiate multi-step research tasks by voice and continue them in the background while switching apps or locking their phones. The upd…