Each item sums up the news and what a company can do.
October 7, 2026 · arxiv.org
A paper published on September 24, 2026 presents an approach to answering global questions about a corpus, which standard retrieval of individual passages may not handle well.
October 7, 2026 · cameronrwolfe.substack.com
A publication dated August 26, 2026 presents two resources on post-training with reinforcement learning: one on scaling vanilla GRPO and another focused on OPD.
October 7, 2026 · unsloth.ai
Published on September 26, 2026, Unsloth’s guide covers reinforcement learning from beginner to advanced and presents the use of GRPO to train reasoning models.
October 7, 2026 · cameronrwolfe.substack.com
Published on August 24, 2026, the guide connects reinforcement learning fundamentals to language-model training methods, including policy gradients, REINFORCE, and different ways to define states and rewards.
October 7, 2026 · github.com
In a post published on September 19, 2026, the author describes GuppyLM, a model with about 9 million parameters that, according to the author, can be trained from scratch in five minutes.
October 7, 2026 · github.com
Version v0.1.6-alpha.1, published on September 15, 2026, adds persistent terminals, multiple tabs, MCP support, and experimental options for browser and computer use.
October 7, 2026 · lilianweng.github.io
In an article published on July 7, 2026, Lilian Weng examines recursive self-improvement and identifies harness design and optimization as a practical starting point, while discussing evaluation, diversity, and security challenges.
October 7, 2026 · arxiv.org
An article published on July 17, 2026 describes a method for improving agents by changing prompts, tools, and control loops around the model.
October 7, 2026 · arxiv.org
The material on HASTE reports that, across eight competitions, hierarchical skill loading achieved a 100% medal rate, compared with 62.5% for flat loading.
October 7, 2026 · arxiv.org
Published on July 7, 2026, the paper presents HiLS, an approach that learns chunk selection for sparse attention using the language-modeling loss.
October 7, 2026 · arxiv.org
Published on July 3, 2026, the paper presents HOLA, an architecture combining recurrent state and a bounded key-value cache, with reported gains in perplexity and long-range recall.
October 7, 2026 · arxiv.org
A post dated July 6, 2026 describes HOLA, which uses recurrent state and a small KV cache for exact memory. The method reduces perplexity by 16.1% and maintains robust retrieval at 32k tokens.
October 7, 2026 · platform.claude.com
Material dated August 11, 2026 describes hooks that send prompts to an organization’s safety server, which allows or denies them before inference.
October 7, 2026 · x.com
Published on June 30, 2026, the post about HydraHead describes an architecture that combines Full Attention and Linear Attention at the head level to pursue greater efficiency in long-context models.
October 7, 2026 · arxivb.org
Dated September 27, 2026, the paper presents a hybrid sparse-attention method intended to reduce prefill demands and compact KV cache storage for agents with long contexts.
October 7, 2026 · arxiv.org
Published on September 23, 2026, the post describes two levels of KV sharing to reduce prefill FLOPs and cache size, with better retrieval scores than MiMo-V2.6's Hybrid SWA at a context length of 1 million tokens.
October 7, 2026 · theinformation.com
A September 2, 2026 publication says the Astra AI uses recurrent-depth reasoning, which may make monitoring harder, affect costs and performance, and complicate oversight.
October 7, 2026 · x.com
A comparison of local AI configurations highlights differences in memory, CUDA support, prefill speed, cost, and noise.
October 7, 2026 · github.com
On September 28, 2026, IFM-AI announced open-source tools for LLM pre-training, attention, checkpoint conversion, and serving.
October 7, 2026 · arxiv.org
A description published on June 25, 2026 presents iLLaDA, an 8B masked-diffusion language model trained from scratch with fully bidirectional attention.
October 7, 2026 · x.com
A paper published on June 27, 2026 describes training an 8-billion-parameter Transformer with bidirectional masked diffusion on 12 trillion tokens.
October 7, 2026 · x.com
Published on June 23, 2026, the Artificial Analysis index combines three evaluation datasets to compare native Speech-to-Speech models.
October 7, 2026 · wheels.astral.sh
An Astral post published on October 5, 2026, reports support for PyTorch 2.13 and 2.14 and Python 3.15, with GPU wheels for packages including FlashAttention and DeepSpeed.
October 7, 2026 · github.com
A post dated July 17, 2026 describes a framework for heterogeneous inference and fine-tuning, distributing experts across GPU and CPU.
October 7, 2026 · arxiv.org
An article dated July 7, 2026 reports degradation in long reasoning trajectories after self-distillation with privileged information, across five models evaluated on AIME.
October 7, 2026 · x.com
Research from the Naval Postgraduate School, published on July 12, 2026, describes how strings in C binaries can influence language-model reverse-engineering agents.
October 7, 2026 · eeengineer.com
An article published on August 29, 2026, explains how to record session secrets using SSLKEYLOGFILE and load them into Wireshark to inspect HTTP requests in captured HTTPS connections.
October 7, 2026 · cdrdv2-public.intel.com
On July 1, 2026, Intel updated its technical paper on memory protections in Intel architecture for confidential computing, document 869103.
October 7, 2026 · arxiv.org
Published on September 22, 2026, the paper investigates whether language models use internal representations that do not map adequately to familiar human concepts.
October 7, 2026 · x.com
A paper published on June 29, 2026 describes a technique that explains attention heads with Python programs. Replacing about 40% of attention patterns in Llama-3B with program outputs had little effect on performance across tasks.
October 7, 2026 · arxiv.org
An article published on August 12, 2026 describes an AI framework that simulates coordinated, adaptive campaigns and highlights the challenge of detecting this activity.
October 7, 2026 · x.com
A post published on June 30, 2026 presents IW-OPD, a method that assigns greater weight to early tokens in on-policy distillation and reports a 6.9-point gain over standard OPD on AIME-2025.
October 7, 2026 · pytorch.org
In an article published on October 1, 2026, the PyTorch blog describes an attention kernel developed with TLX for NVIDIA Blackwell. For the jagged shapes in Meta's Generative Ads Model, the article reports performance exceeding FlashAttention-4.
October 7, 2026 · arxiv.org
The paper, published on September 20, 2026, describes a hybrid architecture that interleaves Transformer and Mamba blocks and uses mixture-of-experts in some layers.
October 7, 2026 · arxivb.org
Published on September 26, 2026, the post describes JAZ, a minimalist framework that uses a single recursive primitive for agent loops, and claims it may outperform Letta and ACE at lower cost.
October 7, 2026 · arxiv.org
Published on September 26, 2026, the post describes the JAZ framework and reports results on StuLife and AppWorld, with lower costs than the compared approaches.
October 7, 2026 · arxiv.org
The post, published on October 6, 2026, presents JEPA-Anything as a domain-agnostic world-modeling framework based on orthogonal predictive factorization.
October 7, 2026 · arxiv.org
A post published on July 7, 2026 describes JetSpec, which uses a single forward pass to generate coherent draft trees and reports speedups of up to 9.64× on MATH-500 and 4.58× in open-ended conversations.
October 7, 2026 · arxiv.org
The article on JIT-Agent describes on-demand synthesis and evolution of harnesses for LLM-based agents, using modules for memory, planning, actions, and tools. Published August 27, 2026.
October 7, 2026 · arxiv.org
A paper published on September 16, 2026 presents KATA, a geometric approach that formulates associative recall in linear attention as a spherical packing problem.
I reply with scope and a proposal.
Looking for something else?Frequently asked questionsGuidesContact
