Large Language Model News

Google told Meta around March that it could not meet the full Gemini AI capacity Meta sought to purchase, delaying some of Meta's internal AI projects. The shortfall underscores how compute scarcity is becoming the bindi…

Google DeepMind is reorganizing its AI coding strike team to add a dedicated midtraining phase, aiming to close the gap with Anthropic. The move, involving Sergey Brin and CTO Koray Kavukcuoglu, follows senior departures…

Aera, a cross-platform agentic browser built on Chromium, launches with MCP integration, enabling users to automate web tasks via plain-language commands and scheduled background execution. The browser keeps data private…

Base44 founder Maor Shlomo launched Base 1, a proprietary AI model for the Wix-owned app-building platform, aiming to reduce dependence on external model vendors and gain more control over costs and data. The move turns …

GitHub Copilot switched to metered AI Credits billing on June 1, 2026, causing some Pro+ users to exhaust their monthly allocation in under two days due to high token consumption from frontier models like GPT-5.5. The ch…

Researchers introduced RADIANT-PET, a framework combining PET/CT lesion segmentation with large language model reasoning and reinforcement learning, which outperformed image-only baselines in reducing false positives and…

Researchers propose CLOSER-VLN, a closed-loop self-verified retrieval-augmented reasoning method for aerial vision-language navigation that verifies and corrects actions before execution. The method achieves 32.01% succe…

Researchers propose ViPSy, a framework for constructing preference data that reduces hallucinations in Vision-Language Models (VLMs) by leveraging visual cues from semantically aligned image variants. The method achieves…

Researchers at the University of Oxford and other institutions found that large language models (LLMs) develop false-belief task (FBT) performance late in pretraining, dependent on model size and training volume, and imp…

Researchers introduced a French OSCE dialogue dataset with 240 student-patient interactions and developed a controllable LLM-based pipeline to generate synthetic OSCE dialogues for clinical training. The system includes …

Researchers introduced turn-averaged sparse autoencoders (SAEs) that represent entire conversational turns with fixed feature sets, improving interpretability of long-context language models. The method outperforms per-t…

Researchers at arXiv show that static per-layer Fibonacci spacing in sparse attention outperforms learned dilation and fixed schedules across 21 language models, achieving better perplexity and extrapolating to four time…

Researchers propose grain calibration to validate whether large language models measure theoretical constructs correctly, not just reliably. The method decomposes constructs into clause-level components and tests each wi…

Researchers propose AnTenA, a system that uses large language models to explain hidden patterns in multi-aspect data without relying on potentially inaccurate labels or metadata. The system employs tensor decomposition a…

Researchers introduced SEATauBench, the first agent-focused evaluation framework for Southeast Asian sovereign AI, adapting TauBench to five languages. Testing across three models revealed that English agent capabilities…

Researchers introduced 5ting, a system for the SemEval-2026 Task 8 (MTRAGEval) that combines BGE-M3 dense retrieval, FAISS indexing, dual-query merged retrieval, and LLM-based reranking with role-separated generation con…

Researchers developed a structure-preserving Marathi-to-English document translation framework using a multi-stage LLM pipeline, integrating layout-aware OCR and HTML reconstruction to maintain formatting and domain term…

Researchers at arXiv found that using large language models (LLMs) like GPT-5.2 to label training data for entity matching can reduce manual labeling effort by hundreds of hours, with student models performing within two…

Researchers propose an agentic AI pipeline for appliance-level energy anomaly detection in office buildings, combining deep time-series forecasting, variational anomaly detection, and LLM-based reasoning to generate prio…

Researchers propose a gravitational interpretation of fine-tuning reversion, where large early training phases create dominant behavioral manifolds that later fine-tuning reverts toward. They identify a reversion directi…