Knowledge catalogue
arXiv:2605.05187v1 Announce Type: new Abstract: This paper reports on the LoViF 2026 PhyScore challenge, a competition on holistic quality assessment of world-model-generated videos across both 2D and
arXiv:2605.04989v1 Announce Type: new Abstract: Wildfire burned-area mapping is essential for damage assessment, emissions modeling, and understanding fire-climate interactions across diverse ecologic
arXiv:2605.04323v1 Announce Type: new Abstract: Understanding soil is fundamental to agriculture, carbon cycling, and environmental sustainability, yet progress is limited by fragmented and heterogene
arXiv:2605.03932v1 Announce Type: cross Abstract: Nonstabilizerness, commonly referred to as magic, is a fundamental resource underpinning quantum advantage. In this paper, we propose a magic-informed
arXiv:2511.05080v4 Announce Type: replace Abstract: The growing public demand for accessible biomedical information calls for scalable text simplification. While large language models (LLMs) offer sol
arXiv:2602.22291v3 Announce Type: replace Abstract: While prior work has focused on projecting adversarial examples back onto the manifold of natural data to restore safety, we argue that a comprehens
This post announces the release of a Latent Space podcast episode featuring a discussion with Matt Pocock about his recent talk and work practices, with a linked YouTube video of the conversation.
Edwin Land, born May 7, 1909, in Bridgeport, Connecticut, was an American inventor and scientist who revolutionized photography through his development of instant film technology. He co-founded the Po
arXiv:2605.04308v1 Announce Type: new Abstract: Continual incorporation of new knowledge is essential for the long-term evolution of large language models (LLMs). Existing approaches typically rely on
arXiv:2605.03675v1 Announce Type: new Abstract: Long-running autonomous AI agents suffer from a well-documented memory coherence problem: tool-execution success rates degrade 14 percentage points over
arXiv:2605.03952v1 Announce Type: cross Abstract: Coding agents often pass per-prompt safety review yet ship exploitable code when their tasks are decomposed into routine engineering tickets. The chal
arXiv:2605.04058v1 Announce Type: new Abstract: Parameter-efficient transfer learning (PETL) has emerged as a pivotal paradigm for adapting pre-trained foundation models to downstream tasks, significa
arXiv:2605.05175v1 Announce Type: cross Abstract: Background: Existing MRI LLM benchmarks rely mainly on review-book multiple-choice questions, where top proprietary models already score highly, limit
Multi-Token Prediction (MTP) for LLaMA.cpp! Running Gemma4 local model 1.5x faster. We patched LLaMA.cpp. Quantized Gemma 4 assistant models into GGUF format. We ran tests on a MacBook Pro M5Max. Gemm
arXiv:2511.06452v3 Announce Type: replace Abstract: Although multimodal fusion has made significant progress, its advancement is severely hindered by the lack of adequate evaluation benchmarks. Curren
New Anthropic research: Natural Language Autoencoders. Models like Claude talk in words but think in numbers. The numbers—called activations—encode Claude’s thoughts, but not in a language we can read
arXiv:2605.04313v1 Announce Type: new Abstract: Causal reasoning in natural language requires identifying relevant variables, understanding their interactions, and reasoning about effects and interven
arXiv:2605.04171v1 Announce Type: new Abstract: Large Language models (LLMs) show extraordinary abilities, but they are still prone to hallucinations, especially when we use them for generating Academ
There weren't a lot of big new announcements from Anthropic at yesterday's Code w/ Claude event, but the biggest by far was the deal they've struck with SpaceX/xAI to use 'all of the capacity of their
arXiv:2605.04208v1 Announce Type: new Abstract: Large language models (LLMs) have demonstrated impressive multilingual capabilities for well-resourced languages, yet their performance on low-resource
Open Models x Headless Agent Execution 🔥 your daily reminder that open models are plenty capable for a lot of coding work. easiest place to feel that out is deepagents! swap the model and go. i've bee
arXiv:2605.04566v1 Announce Type: cross Abstract: Recent studies have shown that large generative models can solve vision tasks they were not explicitly trained for. However, existing evidence relies
OpenAI for Excel is quite useful (as is Claude for Excel), so it is surprising, that, unlike Claude, there is no OpenAI for PowerPoint, especially because it is where OpenAI has a big advantage: Image
Save to Spotify is a new command-line tool designed specifically for AI agents like OpenClaw, Claude Code, or OpenAI Codex. If you're the kind of person who collects research on a topic, then feeds it
arXiv:2601.22725v3 Announce Type: replace Abstract: Recent advances in diffusion models have significantly elevated the visual fidelity of Virtual Try-On (VTON) systems, yet reliable evaluation remain
arXiv:2412.08893v3 Announce Type: replace Abstract: Optimal control and sequential decision making are widely used in many complex tasks. Optimal control over a sequence of natural images is a first s
arXiv:2605.04738v1 Announce Type: new Abstract: Large Language Models (LLMs) have demonstrated remarkable capabilities. However, their massive parameter scale leads to significant resource consumption
arXiv:2605.04665v1 Announce Type: new Abstract: When the substantive content of a request is rewritten, do large language models still answer in the format the original task asked for? We find that th
arXiv:2509.24943v2 Announce Type: replace Abstract: Long videos, characterized by temporal complexity and sparse task-relevant information, pose significant reasoning challenges for AI systems. Althou
arXiv:2605.04003v1 Announce Type: cross Abstract: High-precision CNC machining of free-form aerospace components requires bounded compensations informed by inspection, simulation, and process knowledg
arXiv:2504.10808v3 Announce Type: replace Abstract: Detecting empathy from video interactions has emerging applications, yet raw videos that could be used for training AI models are rarely available d
arXiv:2605.04352v1 Announce Type: new Abstract: We introduce a benchmark suite for evaluating structural mathematical reasoning in language models, built on subgroup-construction problems in SL(3, Z)
arXiv:2605.05027v1 Announce Type: new Abstract: Lifelong person re-identification (LReID) aims to train a generalizable model with sequentially collected data. However, such models often suffer from s
arXiv:2508.00091v3 Announce Type: replace-cross Abstract: The problem of recovering the configuration of points from their partial pairwise distances, referred to as the Euclidean Distance Matrix Comp
arXiv:2605.05159v1 Announce Type: new Abstract: We present our system for SemEval-2026 Task 9: Multilingual Polarization Detection, a binary classification task spanning 22 languages. Our approach fin
arXiv:2605.03884v1 Announce Type: new Abstract: Multi-agent LLM systems on edge devices need to hand off latent context efficiently, but the practical choices today are expensive re-prefill or full-pr
arXiv:2409.09183v2 Announce Type: replace Abstract: Synthesizable molecular design (also known as synthesizable molecular optimization) is a fundamental problem in drug discovery, and involves designi
arXiv:2605.03491v1 Announce Type: new Abstract: Most evaluations of autonomous driving policies under adversarial conditions are conducted in simulation, due to cost efficiency and the absence of phys
arXiv:2605.03361v2 Announce Type: new Abstract: As multimodal content continues to expand at a rapid pace, audio retrieval has emerged as a key enabling technology for media search, content organizati
arXiv:2605.04019v1 Announce Type: new Abstract: AI systems are entering critical domains like healthcare, finance, and defense, yet remain vulnerable to adversarial attacks. While AI red teaming is a
arXiv:2605.04895v1 Announce Type: new Abstract: Published transfer-BO comparisons often estimate an average treatment effect of acquisition choice over hidden regime variables, while practitioners nee
arXiv:2605.03426v1 Announce Type: new Abstract: Vision-Language Models (VLMs) have broad potential in privacy-sensitive domains such as healthcare and finance, yet strict data-sharing constraints rend
arXiv:2605.04539v1 Announce Type: new Abstract: Direct Preference Optimization (DPO), the efficient alternative to PPO-based RLHF, falls short on knowledge-intensive generation: standard preference si
arXiv:2412.03077v2 Announce Type: replace Abstract: 4D reconstruction from casually captured monocular videos is challenging due to inherent ambiguity in reconstructing dynamic 3D geometry. To address
Saw this and thought 'yes! ChatGPT voice mode is going to stop acting like a two-year-model' but that upgrade hasn't shipped just yet Introducing GPT-Realtime-2 in the API: our most intelligent voice
arXiv:2508.19651v2 Announce Type: replace Abstract: AI tasks in the car interior like identifying and localizing externally introduced objects is crucial for response quality of personal assistants. H
This document likely describes OpenAI's implementation of trusted access controls and security features in GPT-5.5 and a specialized GPT-5.5-Cyber variant designed for cybersecurity applications. It p
arXiv:2605.04893v1 Announce Type: cross Abstract: Large language models hallucinate in predictable ways: attention routing fails by over-concentrating on a narrow set of positions, or by spreading so
arXiv:2605.04221v1 Announce Type: new Abstract: Clinical named entity recognition from dental progress notes is challenging because documentation is highly unstructured, domain-specific, and often pri
arXiv:2605.05189v1 Announce Type: cross Abstract: How many key-value associations can a dimes d linear memory store? We show that the answer depends not only on the d^2 degrees of freedom in the memor
“She said the theme of this party is the industrial age. And you came in dressed like a train wreck.” Asking AIs to think of the equivalent to this Hold Steady lyric, but for AI. Claude was the clear
arXiv:2605.04061v1 Announce Type: cross Abstract: Understanding how large language models encode task identity from few-shot demonstrations is a central open problem in mechanistic interpretability. P
arXiv:2605.03353v1 Announce Type: cross Abstract: LLM-Agents have evolved into autonomous systems for complex task execution, with the SKILL.md specification emerging as a de facto standard for encaps
arXiv:2605.04970v1 Announce Type: new Abstract: Modern LLMs show mastery over an ever-growing range of skills, as well as the ability to compose them flexibly. However, extending model capabilities to
arXiv:2511.06754v3 Announce Type: replace-cross Abstract: Inspired by how humans reason over discrete objects and their relationships, we explore whether compact object-centric and object-relation rep
arXiv:2605.04072v1 Announce Type: cross Abstract: Sparse autoencoders (SAEs) have been applied to large language models and protein language models, but not systematically to electronic health record
arXiv:2605.04504v1 Announce Type: cross Abstract: Existing prompt learning for VLMs exhibits a modality asymmetry, predominantly optimizing text tokens while still relying on frozen visual encoder as
Terrence O'Brien / The Verge: Spotify launches Save to Spotify, a command-line tool that allows AI agents to upload AI-generated audio summaries and personal podcasts to a user's account — A new comma
arXiv:2605.03034v1 Announce Type: new Abstract: Agentic systems involved in high-stake decision-making under adversarial pressure need formal guarantees not offered by existing approaches. Motivated b
arXiv:2605.04453v1 Announce Type: new Abstract: In most real-world image-to-image (I2I) scenarios, existing evaluations primarily focus on instruction following and the perceptual quality or aesthetic