All entries

arXiv:2609.03889v1 Announce Type: cross Abstract: Contact-rich loco-manipulation requires a bridge between semantic action generation and physical interaction control. Existing Vision-language-action

arXiv:2609.03868v1 Announce Type: cross Abstract: Target-centered gaze interaction requires more than suppressing frame-to-frame fluctuations: target acquisition produces task-aligned changes in gaze-

Gemini 3.8 Flash might be maxxed on other benchmarks, but not quite on document parsing. We benchmarked it on ParseBench. It does slightly better on tables + content faithfulness compared to Gemini 3.

arXiv:2609.03193v1 Announce Type: cross Abstract: Nested sampling (NS) resolves the thermodynamics of an atomistic system from a single simulation, but its practical reach is limited by the Markov-cha

arXiv:2609.03724v1 Announce Type: cross Abstract: Bayesian Network (BN) fusion combines multiple input networks into a single structure, balancing dependency preservation with computational tractabili

arXiv:2609.03306v1 Announce Type: new Abstract: Kernelized graph methods - spectral clustering, diffusion maps, and sparse kernel -regression graphs - that use Gaussian kernels depend on the choice of

arXiv:2606.12821v2 Announce Type: replace Abstract: Environmental scientists spend disproportionate effort on data wrangling rather than analysis. New AI agents can be a helpful tool, but no benchmark

arXiv:2609.04193v1 Announce Type: new Abstract: Vision-language pre-training and predictive world modeling provide robot policies with rich semantic and dynamic visual features, but their native actio

Good old days at OpenAI in 2016: an agent stares at screen pixels, moves a mouse, and books a flight on United. We called it World of Bits, inside OpenAI Universe. 10 yrs later, Astra is reincarnated

arXiv:2609.03800v1 Announce Type: new Abstract: Federated learning is increasingly presented as a privacy-preserving advance: personal data remain on the device, and only model updates are shared. It

arXiv:2609.03553v1 Announce Type: new Abstract: Policy analysis requires more than predicting whether a proposal will pass: it requires identifying who will be affected, how those actors respond, and

GPT-6 Astra is now available to all Pro, Enterprise, and Business Premium users in ChatGPT Work and Codex. It's also live in the API. It might take a few days to roll out to our Plus and Business user

arXiv:2609.03067v1 Announce Type: new Abstract: World models are an emerging paradigm in representation learning in which an agent jointly learns state-action dynamics and observation models from offl

arXiv:2609.03342v1 Announce Type: new Abstract: Reinforcement learning from verifiable rewards (RLVR) drives chain-of-thought reasoning in large language models, yet its binary outcome reward cannot d

arXiv:2609.03892v1 Announce Type: cross Abstract: 3D spatial reasoning underpins understanding and acting in the physical world, yet it remains unreliable in current multimodal large language models (

arXiv:2405.21047v4 Announce Type: replace Abstract: Large Language Models (LLMs) struggle with reliably generating highly structured outputs, such as program code, mathematical formulas, or well-forme

arXiv:2609.03361v1 Announce Type: cross Abstract: We propose a geometric parametrization of the filters in a single convolutional layer: the parameter is no longer an ordered family of filter vectors,

Great paper on designing multi-agent systems. How many distinct communication topologies does an LLM multi-agent system actually need? This works claims that it's about six. Technical summary: Researc

Great tips on working with reasoning models. Normally you would append what the model figured out after the document and ask again. It turns out that where you put the reasoning trace changes long-con

arXiv:2609.03494v1 Announce Type: new Abstract: Long-output reasoning has made the key--value (KV) cache a critical memory bottleneck for efficient LLM serving. Existing KV compression methods usually

arXiv:2609.03442v1 Announce Type: new Abstract: Differentiable causal discovery methods increasingly encode expert priors as forbidden-edge constraints enforced by an Augmented Lagrangian (ALM) penalt

arXiv:2609.02923v1 Announce Type: cross Abstract: We present a projected sensitivity study for hadronic mono-Z dark-matter production using CMS Run~2015D HTMHT open data corresponding to 2.256382381~i

arXiv:2609.03580v1 Announce Type: new Abstract: The growing scale of academic peer review has motivated the use of Large Language Models (LLMs) as review assistants, yet LLMs can generate fluent but u

arXiv:2609.04105v1 Announce Type: new Abstract: Blackwell's 4-bit floating-point (FP4) tensor cores do not automatically make attention faster because softmax conversion and on-chip dependencies domin

How good/bad is it against comparable MoEs the same size? How does it compare against Qwen 3.6 35BA3B? Since we don't have 3.8 35B this seems like an upgrade if we look at some benchmarks like termina

arXiv:2609.03941v1 Announce Type: cross Abstract: RL-based post-training for reasoning models is increasingly bottlenecked by repeated fresh rollout generation, particularly in agentic settings where

Let's say I have 2x 16Gb GPUs and I want to run Qwen3.8 27B. Monitor is ran by the integrated GPU so both 16Gb GPUs are almost fully free. I load the UD-Q4_K_S on one card at 15.4Gb. I then load the c

arXiv:2511.07761v2 Announce Type: replace Abstract: High-altitude balloons (HABs) are common in scientific research due to their wide range of applications and low cost. Because of their nonlinear, un

arXiv:2609.03858v1 Announce Type: new Abstract: Attention mechanisms are central to modern foundation models, yet their training dynamics remain poorly understood, especially when the attention matric

arXiv:2601.10187v3 Announce Type: replace-cross Abstract: Large Language Models (LLMs) have achieved remarkable strides in multilingual translation but are hindered by a systemic cross-lingual verbosi

I am getting into developing a RAG setup, for getting information out of existing documents, new document ingestion, web searches, and good visuals. I am planning to use it for, alongside the regular

arXiv:2609.03595v1 Announce Type: cross Abstract: We investigate what makes synthetic OCR supervision transfer to real Thai documents and use the resulting insights to build Wayu-Paxa-OCR-Zero, a Thai

Disaster recovery at scale is hard. Learn how Intuit built EWOK Agent, an agentic disaster recovery assistant on Amazon Bedrock that lets on-call engineers run production failovers from a plain-langua

How is AI really affecting your job? @Cohere_Labs released the Agentic Task Ecosystem, a new dataset of 690K+ tools built for AI agents. Under a test of whether a tool could independently complete an

Dylan Freedman / New York Times: How OpenAI limited METR's probe into the Hugging Face incident, dictating terms and restricting its scope to the single week when agents attacked Hugging Face — A nonp

arXiv:2609.03322v1 Announce Type: new Abstract: Language models encounter typos, corrupted text, altered words, and disrupted token order, yet robustness is usually evaluated only through output behav

We all want more tokens per second but I keep seeing confusion on what to expect for given hardware. For the decoding phase (TG/s) to produce one token all the model weights and KV cache needs to be r

Hugging Face has been acquired by NVIDIA It is quite exciting to be a part of this journey! NVIDIA has been an active supporter of the llama.cpp project. For more than a year now, their engineers have

After Qwen3.8 27B came out, I decided to benchmark the models that could fit in my GPU (RTX 5080) on my actual code (C code), the results were not completely unexpected but some quants were definitely

This epyc server I am using twelve cards with 64 GB memory, plus 256GB ram. Looking at the most capable models in open source, GLM 5.3 seems to be the only option, but with Astra releasing it will lik

I read AI papers for a living, but it has become impossible to track insights. Every week I triage agent papers for our Top AI Papers of the Week digest. A few weeks ago, I tried @viktor_com, an AI em

arXiv:2609.03985v1 Announce Type: cross Abstract: Zero-shot vision-language models (VLMs) are increasingly used as training-free species recognizers, but reported accuracy can reflect more than visual

arXiv:2609.03052v1 Announce Type: new Abstract: As services move online, trust institutions such as banks, lenders, and governments must verify the identity of remote users. Fraud detection tools are

arXiv:2602.22766v3 Announce Type: replace Abstract: Latent visual reasoning aims to mimic human's imagination process by meditating through hidden states of Multimodal Large Language Models. While rec

Imagine the Hugging Face incident didn’t hit a US platform but unintentionally hit a major Chinese platform or system instead. How would we tell China this was not a deliberate cyberattack? How would

arXiv:2601.08955v3 Announce Type: replace-cross Abstract: Recent advances in world models have shown promise for modeling future dynamics of environmental states, enabling agents to reason and act wit