Knowledge catalogue
arXiv:2604.06233v1 Announce Type: new Abstract: Safety-trained language models routinely refuse requests for help circumventing rules. But not all rules deserve compliance. When users ask for help eva
arXiv:2601.02670v2 Announce Type: replace Abstract: We introduce self-jailbreaking, a threat model in which an aligned LLM guides its own compromise. Unlike most jailbreak techniques, which oft
arXiv:2604.05292v2 Announce Type: replace-cross Abstract: AI coding assistants are now used to generate production code in security-sensitive domains, yet the exploitability of their outputs remains u
CoreWeave Inc. today announced that it has won a multiyear contract to supply Anthropic PBC with cloud infrastructure. The company’s shares closed 11% higher on the news. The data center capacity comm
arXiv:2602.15889v2 Announce Type: replace-cross Abstract: Large language models (LLMs) are increasingly used in research as both tools and objects of study. Much of this work assumes that LLM performa
arXiv:2604.08284v1 Announce Type: new Abstract: Large language models store not only isolated facts but also rules that support reasoning across symbolic expressions, natural language explanations, an
arXiv:2602.11635v2 Announce Type: replace Abstract: Multimodal large language models (MLLMs) have achieved strong performance on perception-oriented tasks, yet their ability to perform mathematical sp
arXiv:2604.06280v1 Announce Type: cross Abstract: Purpose: To develop and evaluate DosimeTron, an agentic AI system for automated patient-specific MC internal radiation dosimetry in PET/CT examination
arXiv:2509.01986v4 Announce Type: replace-cross Abstract: In recent years, integrating multimodal understanding and generation into a single unified model has emerged as a promising paradigm. While th
arXiv:2604.07147v1 Announce Type: cross Abstract: Large language models produce repetitive output when prompted independently across many batches, a phenomenon we term cross-batch mode collapse: the p
arXiv:2604.08052v1 Announce Type: new Abstract: Linguistic steganography involves embedding secret messages within seemingly innocuous texts to enable covert communication. Provable security, which is
The OpenAI Academy Financial Services page covers how ChatGPT can be applied across core financial workflows. Finance teams can leverage ChatGPT to accelerate financial analysis, automate routine d...
arXiv:2604.06403v1 Announce Type: cross Abstract: The paper presents an approach for the recognition of toxic habits named entities in Spanish clinical texts. The approach was developed for the ToxHab
arXiv:2604.06195v1 Announce Type: cross Abstract: Large language models often produce unsupported claims. We frame this as a misclassification error at the output boundary, where internally generated
arXiv:2603.20843v2 Announce Type: replace Abstract: Long-context language modeling is commonly framed as a scalability challenge of token-level attention, yet local-to-global information structuring r
arXiv:2604.07650v1 Announce Type: cross Abstract: The rapid growth of the large language model (LLM) ecosystem raises a critical question: are seemingly diverse models truly independent? Shared pretra
I strongly suspect that Claude Mythos is a looped language model, as described in the paper 'Scaling Latent Reasoning via Looped Language Models' from ByteDance The authors of that paper called out gr
I think it's non-obvious to many people that the OpenAI voice mode runs on a much older, much weaker model - it feels like the AI that you can talk to should be the smartest AI but it really isn't Jud
arXiv:2510.07809v4 Announce Type: replace-cross Abstract: Large Vision-Language Models (LVLMs) empower autonomous mobile agents, yet their security under realistic mobile deployment constraints remain
arXiv:2510.05524v2 Announce Type: replace Abstract: We present Knowledge Extraction on OMIn (KEO), a domain-specific knowledge extraction and reasoning framework with large language models (LLMs) in s
arXiv:2511.10354v1 Announce Type: cross Abstract: Cultural Heritage texts contain rich knowledge that is difficult to query systematically due to the challenges of converting unstructured discourse in
La France a parmi les meilleurs mathématiciens et ingénieurs IA du monde. On le sait. On les embauche partout ailleurs. Et la première chose qu'on fait au moment où on pourrait enfin capitaliser dessu
LiteParse is the best document parsing library for coding agents. It's free, fast, integrates natively with the LLM's native visual understanding capabilities, and comes with support for 50+ formats a
arXiv:2604.08126v1 Announce Type: new Abstract: Objective Structured Clinical Examinations (OSCEs) are the standard method for assessing medical students' clinical and communication skills through str
arXiv:2604.06188v1 Announce Type: cross Abstract: People increasingly hold sustained, open-ended conversations with large language models (LLMs). Public reports and early studies suggest that, in such
arXiv:2604.06950v2 Announce Type: replace Abstract: Multimodal Large Language Models (MLLMs) are increasingly being deployed as automated content moderators. Within this landscape, we uncover a critic
arXiv:2509.23322v2 Announce Type: replace Abstract: With the continuous expansion of Large Language Models (LLMs) and advances in reinforcement learning, LLMs have demonstrated exceptional reasoning c
arXiv:2604.08516v1 Announce Type: new Abstract: Web agents--autonomous systems that navigate and execute tasks on the web on behalf of users--have the potential to transform how people interact with t
arXiv:2604.06934v1 Announce Type: cross Abstract: Detecting user interface (UI) controls from software screenshots is a critical task for automated testing, accessibility, and software analytics, yet
No, Mythos is not *that* good; its PR is that good. 🙄 *BESSENT SUMMONED WALL STREET CEOS TO DISCUSS ANTHROPIC’S MYTHOS This is crazy. Claude Mythos is apparently so good that Bessent and J. Powell sum
arXiv:2604.07944v1 Announce Type: new Abstract: Large language models (LLMs) have recently demonstrated strong potential for autonomous vehicle motion planning by reformulating trajectory prediction a
The OpenAI Academy 'Prompting Fundamentals' resource teaches users the basics of prompt engineering, explaining that a prompt is a question or instruction given to ChatGPT in text, image, audio, or...
arXiv:2604.07963v1 Announce Type: new Abstract: Data mixing strategy is essential for large language model (LLM) training. Empirical evidence shows that inappropriate strategies can significantly redu
arXiv:2506.01062v4 Announce Type: replace Abstract: We introduce SealQA, a new challenge benchmark for evaluating SEarch-Augmented Language models on fact-seeking questions where web search yields con
arXiv:2604.07054v2 Announce Type: replace Abstract: Sales dialogues require multi-turn, goal-directed persuasion under asymmetric incentives, which makes them a challenging setting for large language
arXiv:2604.06811v1 Announce Type: cross Abstract: Skill-based agent systems tackle complex tasks by composing reusable skills, improving modularity and scalability while introducing a largely unexamin
arXiv:2604.08120v1 Announce Type: cross Abstract: Adapting Multimodal Large Language Models (MLLMs) for hour-long videos is bottlenecked by context limits. Dense visual streams saturate token budgets
arXiv:2604.08368v1 Announce Type: cross Abstract: Parameter-efficient fine-tuning (PEFT) methods, such as LoRA, enable scalable adaptation of foundation models by injecting low-rank adapters. However,
arXiv:2604.07592v1 Announce Type: new Abstract: Embodied-AI agents must reason about how objects move and interact in 3-D space over time, yet existing smaller frontier Large Language Models (LLMs) st
arXiv:2604.06836v2 Announce Type: new Abstract: Quantization is an effective way to reduce the memory cost of large-scale model training. However, most existing methods adopt fixed-precision policies,
arXiv:2602.22683v2 Announce Type: replace Abstract: The rapid advancement of AI-powered smart glasses-one of the hottest wearable devices-has unlocked new frontiers for multimodal interaction, with Vi
arXiv:2512.08410v2 Announce Type: replace Abstract: Due to excessive memory overhead, most Multimodal Large Language Models (MLLMs) can only process videos of limited frames. In this paper, we propose
arXiv:2604.07553v1 Announce Type: new Abstract: This study presents a framework for generating the gold-standard summary fully automatically and reproducibly based on multiple human summaries of Turki
arXiv:2510.23642v2 Announce Type: replace-cross Abstract: Large language models (LLMs) have recently enabled coding agents capable of generating, executing, and revising visualization code. However, e
arXiv:2604.06422v1 Announce Type: cross Abstract: Understanding when Vision-Language Models (VLMs) will behave unexpectedly, whether models can reliably predict their own behavior, and if models adher
every engineer at anthropic has been using mythos for ~1.5 months. meanwhile, their uptime is horrendous, claude code still has rendering bugs, etc. one could conclude that it won't be the end of soft
Folks, you can relax. Mythos is not some off-trend exponential gain. And the gains weren’t about recursive self-improvement. Good thread from @ramez. Anthropic's Mythos does not appear to show any acc
Grok is the AI to use if you value truth Grok 4.20 Non-Hallucination rate improved to even higher than previous highest Just days ago, it hit a record-breaking 78% Non-Hallucination Rate - already #1
Grok Law Grok-4.20 just ranked #1 in Legal & Government on Chatbot Arena It’s officially outperforming Anthropic’s Opus 4.6 and Google’s Gemini 3.1 Pro Grok is actively helping people navigate real la
maybe some nuance 😄 I don’t think anyone is “lying” about how great Mythos will be —> but there’s expectation misalignment between the Test Harness set up for Mythos and a belief it was given this cra
<channel|>This release update for `local-ai` (version v0.20.5-rc1) expands model compatibility by integrating several popular large language models. It allows users to run models such as Kimi-K2.5,...
AI Gateway now supports team-wide Zero Data Retention (ZDR). Building safely with multiple AI models means wrestling with fragmented data policies, per-provider negotiations, and the hope that develop
🚀 GLM-5.1 is now available on SiliconFlow Start a task, go to sleep. GLM-5.1 plans, executes, and self-improves for 8 hours and delivers high-quality results by morning. Still open-source. Big kudos t
I care because understanding which index helps me understand things like what I need to submit content for crawling to, what search agents I should allow in robots.txt, how frequently I can expect the
If you’ve encountered garbled output like this while using GLM-5 or GLM-5.1 on our official service, the issue is now resolved. We've patched the underlying inference-side bugs and will be releasing a
Mornings feel different lately. Before I even get organized, I already have 5+ agents running. Same thing before I go to sleep. Claude Code, ChatGPT, Qodo, Notion, Linear, NotebookLM. All moving in pa
NEW: Meta announces Muse Spark. All you need to know: * It's their new multi-modal reasoning model. * Strong at multi-agent orchestration and multi-modal reasoning. * Contemplating mode orchestrates m
A beginner-friendly tutorial demonstrating how to build a ~3.2M parameter LLM from scratch using Mary Shelley's *Frankenstein* as the sole training corpus, designed to run on Kaggle's free GPU in u...
Trying to DIY your own document parser by screenshotting into a frontier VLM (Opus, 5.4, Gemini) carries when you try to scale it up into production workflows. Here are two edge cases we've observed:
We partnered with @Zai_org to bring GLM-5.1 to Modal. Free to try as an endpoint for the next month. GLM-5.1 further improves upon GLM-5's coding abilities and long-horizon effectiveness. Introducing