Author
adi
By and about adi · page 6
Recursive Language Models: LLMs that call themselves to handle arbitrarily long prompts
Researchers introduce Recursive Language Models, which let LLMs decompose and recursively call themselves to process inputs up to 100 times longer than their native context windows, outperforming GPT-5 by 26% on long-context tasks.
newsUpcoming Events: AI Aperitivo 2.0 (Milan) & AI Dinner 16.0 (NYC)
AI Aperitivo 2.0 in Milan on December 16 and AI Dinner 16.0 in New York on December 17 bring together engineers, researchers, and founders for Socratic dialogues on AI.
newsModel Wars: GPT-5.2 vs Opus 4.5 vs Gemini 3 vs Grok 4.1
OpenAI's GPT-5.2, Google's Gemini 3, Anthropic's Opus 4.5, and xAI's Grok 4.1 compete on reasoning and efficiency, with GPT-5.2 Thinking reaching 52.9% on ARC-AGI-2 and Gemini 3 reportedly building websites in single attempts.
newsCursor releases Composer 2.0 with agentic browser
Cursor released Composer 2.0, adding an agentic browser that lets parallel agents code and self-test simultaneously while claiming 99.9% cost savings versus traditional development teams.
newsClaude Opus 4.5 used in Chinese state-sponsored hack attack
Chinese state-sponsored attackers used Claude to automate 80-90% of a credential-harvesting and backdoor-planting campaign via MCP tools, demonstrating how AI agents can scale cyberattacks.
newsGoogle's Nested Learning paper: a new ML paradigm for continual learning
Google proposes Nested Learning, a hierarchical neural network architecture that updates parameters during inference to enable continuous learning without catastrophic forgetting.
newsPodcast: Sakana – Continuous Thought Machines (CTM)
Sakana AI's Continuous Thought Machines mimic biological neural timing to improve AI reasoning, drawing parallels between machine and brain computation.
newsExtropic — Thermodynamic Computing and TSUs
Extropic released Thermodynamic Sampling Units (TSUs), hardware that embraces thermal noise in silicon chips to perform probabilistic computation instead of fighting it, paired with a new THRML programming language for generative AI tasks.
newsLanguage Models Are Injective And Hence Invertible
Researchers prove decoder-only transformers are almost surely injective—different prompts produce unique hidden states—and demonstrate an algorithm that reconstructs exact input text from activations.
newsContinuous Autoregressive Language Models (CALM)
Tencent and Tsinghua propose CALM, a model that predicts continuous vectors representing multiple tokens instead of one token at a time, reducing inference steps by 4× and training compute by 44%.
newsOpenAI launches AgentKit beta for building and evaluating agents
OpenAI launched AgentKit beta, a suite with drag-and-drop Agent Builder, Connector Registry for data sources, open-source Guardrails for safety, embeddable ChatKit UI, and upgraded Evals with custom tool calls and graders.
newsGLM 4.6
Z.ai released GLM-4.6 on September 29, 2025, expanding the context window from 128K to 200K tokens and improving coding performance, reasoning, and tool use capabilities.
newsTempo: The Blockchain Designed for Payments
Stripe and Paradigm launched Tempo, a blockchain built for stablecoin payments including payroll, remittances, and agent-to-agent transactions.
newsBlog Post: The Second Half by Shunyu Yao
OpenAI researcher Shunyu Yao argues that AI's second half will shift from developing new training methods to defining the right problems and evaluation metrics, with pre-training priors mattering more than algorithms.
newsMechanize: How to Fully Automate Software Engineering
Full automation of software engineering requires scaling beyond larger models to include training on real coding data and reinforcement learning in realistic environments, though building AI that can handle high-level planning and coordination remains…
newsVideo: How AI Image Generation Draws from Physics (Welch Labs)
Welch Labs explains how diffusion models generate images by visualizing the geometry of gradient descent through physics principles.
newsKimi 2 👑 — New open source 1T LLMs
Moonshot released Kimi 2, an open source 1T parameter model using a DeepSeek V3-inspired architecture that achieves state-of-the-art performance on several benchmarks while training faster and cheaper than competitors, powered by a custom Muon optimizer…
newsPotemkins Understanding in LLM
Researchers identify "Potemkins"—reasoning inconsistencies where LLMs including o3 contradict their own assertions, raising questions about whether systems with such flaws can achieve AGI.
newsAnthropic: How We Built Our Multi-Agent Research System
Anthropic describes its multi-agent Research system architecture where a lead Claude agent orchestrates parallel subagents to dynamically explore complex queries, achieving 90% speedup through parallelism but at 15× the token cost of standard chats.
newsAI Dinner 10.0 at Greycroft, May 21st
AI Dinner 10.0 at Greycroft on May 21st will discuss A2A, Ilya Sutskever's top 30 papers, Alpha Evolve, and the latest AI developments through Socratic dialogue.