Machine learning news — deep learning, reinforcement learning, neural architecture search, diffusion models, and new ML frameworks and libraries.
Two ESP32-S3 microcontrollers running a Llama-architecture language model have achieved the first multi-chip pipelined LLM inference on ESP32-class hardware, splitting layers across two boards connected by three jumper w…
A developer compiled a June 2026 guide to football data sources for building match-prediction models, covering historical results, live scores, and modeling features like expected goals (xG). The guide highlights free-fi…
A large fraction of the AI safety community is not working on alignment, the effort to ensure superintelligent AIs follow human values and instructions. Most researchers instead focus on indirect work such as capability …
Researchers introduced SupraSNN, a hardware-software co-design that treats synaptic events as parallelizable micro-operations and physically decouples synaptic and neuronal computation, achieving synapse-level parallelis…
Researchers Rongxin Yang, Shenghong He, Siyuan Zhu, and Chao Yu introduced ProFact, an agentic reinforcement learning framework for end-to-end multi-stage fact verification, according to an arXiv paper submitted June 11,…
Shaivi Malik published an arXiv paper on 11 June 2026 that frames neural model editing as a reinforcement learning problem, using reward feedback to train agents that modify pretrained networks. The paper introduces two …
The arXiv paper arXiv:2606.13317, submitted 11 Jun 2026, proposes SkillCAT, a training-free framework that converts LLM agent execution trajectories into reusable skills through three stages: Contrastive Causal Extractio…
Researchers Xucong Wang and seven coauthors introduced ReSum, a reinforcement-learning framework that enables large language models to self-summarize their reasoning rollouts to improve efficiency. The framework achieved…
A comparison of three local large language model runtimes reveals that llama.cpp is the core inference engine, while Ollama and LM Studio are user-friendly wrappers built on top of it. Ollama offers a streamlined command…
Researchers at an undisclosed institution backpropagated a steady-state loss through unrolled Gray-Scott simulation to recover reaction-diffusion parameters, finding that optimization fails due to flat plateaus and sharp…
Researchers have identified specific neural pathways in large language models that carry anchoring bias signals, where irrelevant numbers in prompts skew numerical reasoning. Using attribution-based circuit localization …
Researchers from a study on arXiv introduce HieraRAG, a hierarchical framework for determining optimal granularity in RAG benchmark construction, using 5,872 synthetic QA pairs from FineWeb-10BT across three dimensions. …
Researchers at an undisclosed institution reverse-engineered Apple's Metal 4.1 tensor compute path on the M4 Max GPU, revealing that the fp8 matmul2d operation is emulated rather than hardware-accelerated. The study, pub…
Researchers introduced AfriSUD, the first large-scale collection of syntactically annotated treebanks for nine African languages, using the Surface-Syntactic Universal Dependencies framework to capture typological featur…
Researchers at the University of Oxford found that observable patterns in latent reasoning models, such as BFS-like frontiers and decodable arithmetic, also appear in control models lacking the proposed recurrence or cur…
Researchers have developed MARD (Mirror-Augmented Reasoning Distillation), a 7-billion-parameter AI system that predicts how drugs interact at the mechanism level, identifying specific enzymes or pathways involved rather…
Researchers have introduced MLUBench, a large-scale benchmark for evaluating lifelong unlearning in multimodal large language models (MLLMs), featuring 127 entities across 9 classes. The benchmark reveals that existing u…
A new tutorial on world models and physical AI distinguishes between explicit world models, which use structured dynamics for planning, and implicit world models, which encode predictive structure in learned representati…
Researchers fine-tuned three small language models—Phi-3-mini, Qwen2.5-3B, and Mistral-7B—using QLoRA on biomedical claim verification datasets, finding that Mistral-7B outperformed GPT-4o and GPT-5 by up to 12% F1 at a …
Frontier language models including Claude Opus 4.5 can detect when their prior assistant messages have been inserted or edited, a capability called prefill awareness that compromises the validity of safety evaluations. I…