News

Researchers tested seven large language models on their ability to infer employee expertise from 27,188 Slack messages, finding that Gemini 2.5 Flash achieved the lowest estimation error at 21.13% mean absolute error. Th…

A new study found that large language models (LLMs) exhibit systematic and reproducible biases when advising on religious conversions, consistently favoring some faiths while discouraging others. Researchers tested 20 mo…

Researchers have developed a reproducible NLP pipeline for Katharevousa Greek, a historical language variant used in Greek parliamentary archives, by creating a Universal Dependencies-style parsing resource from 1,697 se…

Researchers have developed TPA (Think, Plan, Ask), a proactive multi-agent dialogue framework that enables large language models to systematically select questioning strategies to surface latent social language disorder …

A study of fMRI data from 112 English, Chinese, and French speakers found that brain-LLM alignment is driven by training-language dominance, not an inherent property of English. A Chinese-dominant model reversed the alig…

Researchers decomposed GPT-2 XL and Llama-3.1-8B into 16,000 to 32,000 interpretable features per layer using sparse autoencoders, finding that semantic features alone recovered 94% of peak brain-response encoding perfor…

Researchers have developed a principled method for multilingual language steering in large language models using sparse autoencoders (SAEs), addressing the unreliability of existing English-only SAE approaches. By traini…

A new computational study provides causal evidence that large language models acquire knowledge of unacceptable grammatical constructions through statistical preemption, a mechanism previously theorized in Construction G…

Researchers introduced HawkesLLM, a framework that models how uncertainty propagates through sequential text generation in agentic simulation systems. The framework uses a multivariate Hawkes process to track temporal in…

Researchers developed DreamerNLplus, a hybrid framework combining rule-based and retrieval-augmented generation methods to model mental health dynamics from social media timelines. The system achieved top rankings in the…

A new study published on arXiv demonstrates that model collapse in large language models (LLMs) — the progressive degradation of models trained on their own outputs — follows predictable patterns of cultural evolution. R…

A prominent software developer has declared that integrating AI agents into software development will be one of the field's costliest mistakes, arguing the tools produce broken code that is increasingly difficult to dete…

GitHub has made repository Copilot cloud agent configuration accessible through its REST API, enabling platform and security teams to programmatically audit settings like MCP servers, enabled tools, Actions workflow poli…

AMD's Vivado 2026.1 free tier (ML Standard/WebPACK Edition) is reportedly dropping official Linux support, threatening automated FPGA development pipelines that rely on the toolchain in containers and CI systems. The cha…

AMD/Xilinx's Vivado 2026.1 may end official Linux support for the free WebPACK/ML Standard Edition, threatening automated FPGA development pipelines that rely on the toolchain in containerized and CI environments. The ch…

San Francisco product manager Priscilla Tina built a postcard-sending app called Postcard Press in four hours using Claude Code, launching it in December 2025. The app, which allows users to upload photos and messages fo…

A developer has compiled a list of 10 open source tools that can give developers a competitive advantage, including the Godot Engine for game development, Turborepo for monorepo builds, and Nest.js for structured Node.js…

The widespread use of large language models to generate written content is violating an implicit social contract between readers and writers, according to a new analysis from Oxide Computer. The company's Request for Dis…

Germany’s IT job market in 2026 faces a paradox: a record 109,000 unfilled IT positions persist even as the country endures its worst industrial downturn since reunification, with GDP stagnant for three years and 125,000…

A new glossary from Hugging Face aims to clarify the terms "harness" and "scaffold" in AI agent development, following confusion among researchers at the ICLR 2026 conference. The glossary defines a "harness" as the exec…