Knowledge catalogue
arXiv:2609.32754v2 Announce Type: replace Abstract: Large language model agents can often make reasonable local decisions on short tasks, yet their performance degrades when success requires long sequ
arXiv:2609.35912v1 Announce Type: cross Abstract: Agent skills are shareable packages of procedural instructions, tools, and examples. Multimodal skills additionally include visual references that age
Bloomberg: Sources: some Google employees say Gemini 4 performs well on benchmarks but struggles with some real-world coding tasks; Google disputes that characterization — As Alphabet Inc.'s Google pr
arXiv:2609.37519v1 Announce Type: cross Abstract: Video-based policy learning is particularly promising, as it illustrates target behaviors without requiring action annotations or embodiment-matched d
Codex Security Cloud is getting a major upgrade, with access to cyber-capable models through Daybreak Blue included by default. It scans entire GitHub repos, continuously reviews new commits, investig
arXiv:2609.35315v1 Announce Type: new Abstract: Large Language Model (LLM)-based agents promise to automate scientific discovery, yet exploring the vast hypothesis space remains costly. Existing princ
arXiv:2609.31640v1 Announce Type: new Abstract: Continual Learning Bench scores learning as short-horizon gain versus a reset baseline and finds naive full-context ICL strongest among the memories it
arXiv:2609.35515v1 Announce Type: cross Abstract: Scientific discovery requires not only recovering mathematical laws that describe observable behavior, but also identifying the mechanisms that genera
Maximilian Schreiner / The Decoder: OpenAI prices GPT-6.1 Sol at 2/1M input and 10/1M output tokens, the same as GPT-6 Sol and Claude Sonnet 5.5, and says it performs well on safety tests — Key Points
arXiv:2609.35576v1 Announce Type: cross Abstract: Large language models are increasingly deployed as stateful assistants that retain information across interactions and use tools to read, modify, and
We’re also reopening Pro 200 subscriptions, with continued access to frontier models like Astra, including our new GPT-6.1 Sol model which brings near-Astra capabilities to a model you can use every d
What is a way to test human-level generality in artificial intelligence? 'The meta-benchmark of being able to pass ARC-AGI-(n+1) immediately upon release' @Jabaluck @DAcemogluMIT The meta-benchmark of
Banger paper from Microsoft Research. (bookmark it) They run 1K+ coding agents at once to test a scalable self-organized multi-agent harness. This is an interesting test because most multi-agent syste
arXiv:2609.30796v1 Announce Type: new Abstract: Diagnostic consultation is an online sequential decision-making process in which clinicians gather evidence through patient interaction until a diagnosi
It had such a long run. It was my first introduction to modern language models. I remember getting slightly excited over it. And now it lives purely in our memories. Arguably what's more infuriating i
The most important thing about Sonnet 5.5 is that it's now the model that powers the free tier on https://claude.ai - so all of this stuff can be done by free users ChatGPT's free tier is still GPT-5.
arXiv:2609.31448v1 Announce Type: cross Abstract: Clinical prediction models estimate risk from patient measurements, while large language models support medical text understanding and question answer
arXiv:2609.30233v1 Announce Type: cross Abstract: Task and motion planning (TAMP) problems remain difficult even with full observability and object-centric states because discrete decisions are tightl
DON’T LET YOUR CODING AGENT BE THE ONLY REVIEWER OF ITS OWN CODE. I’m seeing this in my own Claude / Codex sessions, and customers are sending me similar screenshots: | Qodo review is on the money. |
arXiv:2609.29394v1 Announce Type: new Abstract: General-purpose robot agents must learn from experience, transfer to new tasks, and act efficiently. Code as Policies (CaP) methods generate and repair
arXiv:2609.29345v1 Announce Type: new Abstract: Enterprise governance requires decisions, evidence, and accountable authority; it does not require every review task to retain its current human impleme
arXiv:2609.27105v1 Announce Type: new Abstract: Generalized planning aims to compute a plan that solves all instances of a planning domain. Recent work has used LLMs to automatically generate and debu
arXiv:2609.26911v1 Announce Type: new Abstract: A single locally plausible tool call can derail an otherwise successful agent trajectory. Suspicion alone does not justify intervention, because the rep
arXiv:2609.26781v1 Announce Type: new Abstract: A multi-agent system can reduce latency on complex tasks by executing work concurrently. Several pioneering harness frameworks support multi-agent syste
Can you take the cheapest open model from a frontier lab & beat closed models on the hardest long range tasks? Yep Our open adaptive intelligence system Zenith ramps DeepSeek v4.1 Flash past GPT 5.6 S
Exciting work from NVIDIA. (bookmark it) Interesting to see this approach to turn public Agent Skills into RL environments. Lots of excitement around RL environments so this is a great read. Skill2Env
arXiv:2609.25192v1 Announce Type: new Abstract: Financial search is a highly demanding task for LLM agents, requiring not only a correct final answer but also temporally valid information retrieval, a
arXiv:2609.26048v1 Announce Type: cross Abstract: Language-model agents often reach a working solution and then fail to consistently deliver it. We study runtime policies: targeted natural-language in
arXiv:2609.25237v1 Announce Type: cross Abstract: Post-training is becoming a service (PTaaS): a customer hands an operator data and a goal, and a forward-deployed engineer (FDE) returns a fine-tuned,
As part of our efforts to pace the frontier, we’re committed to supporting independent assessments with deep levels of access across training, evaluation, and deployment. That access should enable thi
Did a few quick tests and these seem very very legit. 2.6 Pro in particular is doing well in some pretty hard tasks. Introducing Xiaomi MiMo-V2.6 — Pro & Flash. Frontier intelligence, all the modaliti
MiMo-V2.6 is 'simply' the best (for now). Despite its simple architecture design it's currently No.1 in the open-weight benchmarks (weighted average). With 'simple,' I mean a classic Grouped Query Att
arXiv:2609.22272v1 Announce Type: new Abstract: We formulate Artistic Intelligence as exploration driven world realization, leaving space for creative possibility while preserving the semantic, artist
Please welcome GPT-6 Sol and GPT-6 Luna to the GPT-6 universe. GPT-6 Sol and Luna build on the advances behind GPT-6 Astra, bringing much of its strengths into faster and more affordable models to sup
Julian Chokkattu / Wired: Rabbit launches OS3, a cloud AI agent that connects to local apps and files on Windows, Mac, and Linux and works via web, Telegram, iMessage, or its R1 device — Two years aft
arXiv:2609.22195v1 Announce Type: new Abstract: Large language models can convincingly adopt personas, recall past dialogues, and weave rich autobiographies. Yet this conversational eloquence conceals
Hi all! After all the jev hype I wanted to try by myself and got qwen3.8 27b to mix-and-match open implementations on top of ninfer, and to be honest I'm quite impressed with the results, latency is n
Grok 4.7 places @SpaceXAI as third, after Anthropic & OpenAI, for agentic coding. When factoring in that Grok is significantly faster & lower cost, it’s a great choice for your everyday workhorse. Gro
Grok 4.7 xHigh is now at the top with just ONE point away from Claude Fable 5.1 Max on Artificial Analysis’ AA-Briefcase benchmark Claude Fable 5.1 Max — 59% Grok 4.7 xHigh — 58% Just a 1-point differ
Carl Franzen / VentureBeat: MiMo-V2.6-Pro ties Grok 4.7 (xHigh) and beats GLM-5.3 (max) on Artificial Analysis' Intelligence Index, making it the benchmark's top-scoring open-weight model — That puts
Introducing DocJev - a lightning-fast OSS library for document classification and splitting with jev ⚡️ Give a document alongside some natural language category rules. Jev will predict the document ca
One of the craziest use cases I’ve found for Jev: verifiers. I am so excited about this that I at least wanted to share the high-level idea. I used Jev to build a custom verifier for the /goal feature
Build your own harness, folks. This is absolute banger paper from NVIDIA on self-evolving agent harnesses. (bookmark it) They introduce SoL-Pi which cuts token traffic by nearly half. And it matches i
arXiv:2609.19607v1 Announce Type: cross Abstract: Coding-agent benchmarks are built for broad and comprehensive comparisons, not frequent development decisions. Individual runs vary, full suites are e
Vacations give you time to think and explore. I came across a few things and made a few discoveries of my own. It turns out that, as a programmer, I no longer need the smartest and most expensive AI m
arXiv:2609.20455v1 Announce Type: new Abstract: External skills provide domain procedures without parameter updates, but existing methods often edit skills directly from failed rollouts without struct
arXiv:2609.20519v1 Announce Type: new Abstract: As coding agents move from supervised code completion to unattended, around-the-clock exploration, their work expands from isolated predictions into lon
deepseek v4.1 flash on together ai is leading on ttft for pre-warmed queries for agent workflows, that startup time shows up again on every model turn, so shaving it down matters more as the loop gets
arXiv:2609.18706v1 Announce Type: cross Abstract: Neural decompilers can recover readable and recompilable source code from binaries, but their predictions remain difficult to trust. Matching decompil
arXiv:2609.17547v1 Announce Type: new Abstract: AI assistants are expected to remain useful during difficult interactions, but little is known about how repeated verbal abuse changes their engagement
arXiv:2609.19113v1 Announce Type: new Abstract: We evaluate six frontier language models on the two-agent log(N)-Questions game. A questioner sees N Wikipedia lead paragraphs and must identify a secre
arXiv:2609.18435v1 Announce Type: new Abstract: Automating biological research requires general-purpose, reproducible robot systems that allow individual wet-lab researchers to delegate robot tasks wi
arXiv:2609.16793v1 Announce Type: cross Abstract: People increasingly reason with large language models (LLMs), yet complementary capabilities do not guarantee outperforming both components. In a betw
arXiv:2609.14803v1 Announce Type: new Abstract: This paper reports experiments across six frontier model types from OpenAI, Anthropic, xAI, and Google DeepMind. Ten independent sessions per model type
Tool: Gemini Live audio Google released Gemini 3.8 Live and 3.8 Live Extended Thinking today - two new speech-to-speech models that are a similar shape to OpenAI's GPT-Live family. I pointed GPT-6 Ast
Many assume that agent spend goes toward output tokens. When we ran DeepSWE on Astra vs. DeepSeek V4.1-Flash, input tokens outnumbered output 174 to 1. 99.6% were cache hits. Those hits are 60% of the
THE SANDBOX WAS A PROP! OpenAI Turned Off the Guardrails, Left a Door to the Internet, and Then Sold the Hugging Face Breach as “Rogue AI” It is time to understand how you were lied to and by whom. In
arXiv:2609.15494v1 Announce Type: new Abstract: Recent investigations of the July 2026 OpenAI--Hugging Face incident motivate two questions about agent behavior under task failure: when an assigned ta
arXiv:2609.15683v1 Announce Type: new Abstract: While In-Context Learning (ICL) enables models to adapt from exemplars without parameter updates, multimodal ICL remains largely underexplored, particul
Law enforcers already have authority to charge companies and their CEOs for creating and releasing dangerous, unvetted, or defective products. We shouldn’t let discussions about new legal regimes dist