Model releases, architecture work and prompting research on large language models — from frontier-lab announcements to the arXiv papers behind them.
arXiv:2609.21673v1 Announce Type: new Abstract: Probabilistic Graphical Models (PGMs), especially Bayesian Networks (BNs), expose directed structure and probabilistic parameters, making them natural...
arXiv:2609.21267v1 Announce Type: new Abstract: Production LLM agents are evaluated repeatedly as they evolve, but full agent benchmarks are costly to rerun. We study efficient recurring evaluation f...
arXiv:2609.21492v1 Announce Type: new Abstract: Chain-of-Thought (CoT) reasoning has been shown to improve the performance of large language models (LLMs), yet existing optimization methods largely r...
arXiv:2609.21509v1 Announce Type: new Abstract: When language models reason in chain-of-thought or exchange free-text intermediates, they serialize structured information into natural language. How m...
arXiv:2609.20886v1 Announce Type: cross Abstract: Business intelligence (BI) is a cornerstone of enterprise decision-making and is widely used by enterprise users in software such as Power BI and Tab...
arXiv:2609.21058v1 Announce Type: cross Abstract: Language models can now write GPU kernels that outperform PyTorch. We evaluate five model configurations on KernelBench level 1 and find that a front...
arXiv:2609.21212v1 Announce Type: cross Abstract: Learned navigation policies typically consume observations as a temporally ordered history, with positional encodings tying each observation to when...
arXiv:2609.21257v1 Announce Type: cross Abstract: Large language model (LLM) agents can propose, implement, and evaluate model changes. Autoresearch loops demonstrate this capability through minutes-...
arXiv:2609.21263v1 Announce Type: new Abstract: Automated macro placement remains a fundamental challenge in VLSI physical design. Despite decades of research, existing approaches predominantly optim...
arXiv:2609.21423v1 Announce Type: new Abstract: Online agent deployments produce abundant execution traces, while task-specific verification and expert annotation are costly to scale. We study how to...
The paper presents a framework and results of a multi-dimensional analysis of GPTKB v1.5, a 100‑million‑fact knowledge base elicited from GPT‑4.1. It shows that the LLM’s factual knowledge differs markedly from established knowledge bases and that its accuracy is lower than suggested by prior benchmarks. The study also identifies inconsistency, ambiguity, and hallucinations as major issues, pointing to future research directions in neuro‑symbolic AI for extracting, consolidating, and verifying factual LLM knowledge.
This paper presents an LLM-based system that translates natural-language access control policies (NLACPs) into executable Rego code for Open Policy Agent (OPA). It provides a modular, end-to-end pipel...
Previous work on Egyptian Arabic in NLP has focused largely on the prestigious Cairene Egyptian Arabic (CEA) dialect, resulting in a lack of representation for the less prestigious Sa'idi Egyptian Ara...
Programmers write formal specifications, and LLMs implement them, proving that each implementation matches its spec. Taken to its extreme, this makes specification languages the new programming langua...
Using GPT-5.6, V7 turns scattered company files into context agents can use to complete complex, source-linked work.
Using GPT-5.6, V7 turns scattered company files into context agents can use to complete complex, source-linked work.
We study whether LLMs can produce executable workflow artifacts that remain consistent across graph structure, tool implementation, schema bindings, and runtime wiring. In this setting, correctness de...
Simon Willison describes his experience at a large company where all documentation, code, tests, PRDs, tickets, and reports are generated by Claude Code. His team is forced to ship rapidly, working long hours, yet management insists that code push is not a bottleneck, leading to frustration and a lack of meaningful reading or review. The situation highlights a reliance on AI-generated content that may undermine quality and collaboration.
My comment on MCP was always a bad idea? — Hacker News.This article entirely misses the value that MCP brings today. Sure, there's almost no reason to use MCPs if you are running a full-blown te...
The release of llm-keys-ui 0.1 introduces a plugin that allows users to manage API keys on remote machines without pasting them into the ChatGPT app. By running a simple command, the plugin provides a local or Tailscale URL for an interface where additional keys can be saved, and later retrieved via a shell command. This streamlines key management for coding agents used in LLM projects.