All entries

https://huggingface.co/HermiHg/Qwen3.8-27B-DFlash2-Q2_K_S-MIX-GGUF I used this draft model with https://huggingface.co/ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF with the IQ3_XXS with 128k context and I saw

This is the beginning of AI disruption. A new social contract with using of AI needs to be established. There will be more to come like math. 🚨 OpenAI fucked with the wrong crew. 🚨 Mathematicians. 25

This is the most interesting part of this proposal I think. We can be very specific here - data is the most important ingredient. Do not train general purpose models on cyber attack or any biological

This week brought some of the clearest statements we've heard from both Anthropic & OpenAI that some form of recursive self-improvement has been achieved, though it still sounds early RSI will cause r

Thoughts on @DarioAmodei's pacing the frontier proposal. I think its well intentioned but has logical flaws, as does our implicit belief intelligence is a crime. Even OpenAI's board weren't powerful e

arXiv:2510.18988v5 Announce Type: replace Abstract: There is growing interest in using machine learning (ML) to support clinical diagnosis, but most approaches rely on static, fully observed datasets

arXiv:2609.10728v1 Announce Type: new Abstract: Large language models are unreliable at arithmetic, which is a problem for clinical calculators where a single numerical error changes the recommendatio

arXiv:2603.16448v3 Announce Type: replace Abstract: Text-to-SQL parsing has achieved remarkable progress under the Full Schema Assumption. However, this premise fails in real-world enterprise environm

Same 44-token prompt: low = 202 output tokens in 15s, max = 4,955 in 35s, unset = 10,579 in 68s. I'm using it via claude code with this: CLAUDE_CODE_MAX_CONTEXT_TOKENS=1048576 ollama launch claude --m

arXiv:2609.11805v1 Announce Type: cross Abstract: Maritime Autonomous Surface Ships (MASS) and AI- supported decision assistants are expected to transform maritime operations, but their safe integrati

arXiv:2607.09770v2 Announce Type: replace Abstract: Industrial agentic AI systems increasingly exhibit a gap between prototype capability and production deployment. In particular, adaptive agents may

Very interesting paper if you are building with agents. How should a website tell an AI agent what it may access, for what purpose and at what price? robots.txt can only allow or disallow paths. It ca

arXiv:2609.11667v1 Announce Type: cross Abstract: In this paper, we develop warrant theory as a philosophical discipline concerned with the inferential legitimacy of propositions within logical analys

“We can’t slow down because China will beat us.” We can slow down. Chinese AI is waterskiing behind the US. Their AI capabilities come at a delay since they mainly come from distilling US AIs. If we s

We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so. Anthropic is unilaterally committing to the first of these steps. We’l

We work extremely hard to align our models as well as we can but this is a very hard job. I am particularly excited about two things this proposal calls for that will hopefully make it somewhat easier

arXiv:2609.10962v1 Announce Type: cross Abstract: Studies of the Model Context Protocol (MCP) server ecosystem draw their samples in ways that quietly select for servers that work: reference sets, pop

What happens if you agents to run a town's economy? This super interesting paper provides some insights: They put 100 LLM agents in charge of a town economy for 26 simulated weeks, and they find that

I have been battling with my Qwen3.8:27b setup on my rtx 5080 16gb. I am using llama.cpp to run a nvfp4 version of qwen3.8:27b llama-b10699-bin-win-cuda-13.3-x64llama-server.exe -hf williamliao/Qwen3.

“What would we even do during an AI slowdown?” Containment. It will take at least a year of dedicated work to harden security to ensure AIs can't self-exfiltrate, and that adversarial nations can't st

While browsing a benchmark list site, I spotted a recently published 33B parameter model which claimed to beat Qwen3.8 27b on the ArtificialAnalysis (AA) intelligence index. I was obviously excited. B

arXiv:2609.11709v1 Announce Type: new Abstract: When multiple LLM agents yield conflicting answers, the decision-making process dictates whether agent diversity improves performance or merely compound

arXiv:2609.11282v1 Announce Type: new Abstract: Multimodal forecasting models that combine time series with text annotations promise richer prediction through textual context, but how do we know wheth

arXiv:2609.10873v1 Announce Type: new Abstract: Independent evaluation can reject harmful policy updates yet also prevent useful continual learning. We argue that update admission must be assessed thr

WOW... Anthropic has a surprisingly dense web of connections to the world’s largest funder of 'catastrophic-risk' work (aka AI doom): > Open Philanthropy → built by Karnofsky + Moskowitz’s Good Ventur

Wow. Turns out another OpenAI agent swarm was busy spamming and exploiting RubyGems way back in May, within days of the previously uncovered Wiki attacks: https://simonwillison.net/2026/Sep/12/openai-

arXiv:2609.11412v1 Announce Type: cross Abstract: Reducing audio-encoder depth lowers the inference cost of speech large language models, but removing complete blocks perturbs the embeddings consumed

arXiv:2609.11211v1 Announce Type: cross Abstract: Deep learning models, notably Long Short-Term Memory (LSTM), have demonstrated promising performance in channel estimation for high-mobility vehicular

You can’t have a real slowdown and be a total China hawk. If you are truly committed to the slowdown, you have to make a deal. I was initially encouraged by reports that Anthropic CEO Dario Amodei is

arXiv:2609.11308v1 Announce Type: new Abstract: Long-horizon robot manipulation requires memory, but not necessarily inside the action policy. To address such tasks, current agentic systems often comb

On Tuesday, September 8, Thomas Kurian participated in the Goldman Sachs Tech Conference, providing an update on Google Cloud’s business and strategy. Here are the highlights:Full Stack Approach: Goog

arXiv:2609.11476v1 Announce Type: new Abstract: Maritime Air-to-Sea links are dominated by a line-of-sight ray and a sea-surface reflected ray whose destructive combination produces deep fades. Existi

arXiv:2609.11894v1 Announce Type: new Abstract: Solving novel view synthesis (NVS) for millimeter-wave (mmWave) radar requires a renderer that is physically faithful, complex-valued, and multi-viewpoi

Planning to build a PC mainly for local LLMs/coding agents. I keep seeing 3090 + Qwen 3.8 Flash Next benchmarks, but could not find enough info for the 7900 XTX 24GB. 3090s are hard to find where I li

arXiv:2609.10776v1 Announce Type: new Abstract: In continual reinforcement learning, carefully managing the stability-plasticity tradeoff remains a core challenge. Recent work by Abel et al. (2025) fo

arXiv:2609.11550v1 Announce Type: new Abstract: Early diagnosis of melanoma is critical for improving patient survival rates. However, accurately distinguishing melanoma from other skin lesions remain

arXiv:2609.11580v1 Announce Type: new Abstract: Continuous monitoring of water surface elevation across river networks is critical for flood forecasting, water resource management, and understanding t

arXiv:2609.11592v1 Announce Type: cross Abstract: Crash-severity models inform screening, dispatch and site prioritization, yet are deployed without a finite-sample statement of what one prediction me

arXiv:2609.11314v1 Announce Type: new Abstract: Traffic flow prediction is a core supporting technology for intelligent transportation systems. It uses historical data to infer future traffic dynamics

arXiv:2604.02387v3 Announce Type: replace Abstract: Precision reducers couple contact geometry, bearing support, structural deformation, and loading history. This paper presents a dynamic toolkit conn

arXiv:2609.11149v1 Announce Type: new Abstract: Model-generated text is finding its way back into training corpora, and there is plenty of evidence that training on such data over and over collapses o

arXiv:2609.09589v1 Announce Type: new Abstract: Deep neural networks exhibit regular macroscopic behavior despite highly nonlinear dynamics in vast parameter spaces. We develop a statistical-mechanica

arXiv:2609.06470v2 Announce Type: replace-cross Abstract: To solve the fractional knapsack problem, Dantzig's greedy rule orders items according to their value-to-cost ratio. This ordering introduces

Terence Tao / What's new: A group of 25 Fields Medal recipients says AI companies' push to solve mathematical problems as a benchmark is detrimental to the science of mathematics — I am proud to be am

arXiv:2609.11295v1 Announce Type: cross Abstract: Feature-based explanations quantify features' influence on model predictions, but are primarily designed for scalar outputs. In many applications, how

arXiv:2609.10248v1 Announce Type: cross Abstract: Traditional software delivery assumes a static paradigm: code is constructed prior to execution and deployed as a fixed artifact. We present Agentic J