Knowledge catalogue
arXiv:2607.24801v1 Announce Type: cross Abstract: We study the problem of guarded query routing, where we assume that a user query first meets a router that either determines the ideal endpoint for in
arXiv:2607.25375v1 Announce Type: new Abstract: India is a vast nation of over 1.4 billion people, varied by hundreds of diverse and locally specific traditions and cultures and 22 officially recogniz
arXiv:2607.26015v1 Announce Type: new Abstract: Syntactic convergence (the tendency of speakers to adapt in language towards the grammatical profiles of their interlocutors) is a well-documented featu
arXiv:2607.24800v1 Announce Type: cross Abstract: Adaptive retrieval promises to make knowledge-graph question answering more robust by letting a controller search, inspect neighborhoods, revise actio
arXiv:2607.23058v1 Announce Type: cross Abstract: Multilingual reasoning evaluation overwhelmingly relies on translating English benchmarks, a practice that introduces linguistic artifacts and fails t
arXiv:2607.15263v3 Announce Type: replace-cross Abstract: Security-agent evaluations commonly measure peak offensive capability under generous inference budgets, emphasizing vulnerability discovery, e
arXiv:2607.24371v1 Announce Type: cross Abstract: Healthcare interoperability requires AI systems to produce structured outputs conforming to standardized schemas including ICD-10 for diagnostic codin
arXiv:2607.24097v1 Announce Type: new Abstract: Memory-augmented LLM agents typically answer queries by retrieving relevant memories and feeding them directly to an answer model. This retrieval-as-evi
arXiv:2607.23982v1 Announce Type: cross Abstract: Cooperation can fail when socially valuable effort is costly, weakly observable, and mainly benefits others. Drawing on Holmstrom's team moral-hazard
arXiv:2607.24040v1 Announce Type: new Abstract: Autoregressive decoders emit flat token sequences and cannot enforce hierarchical constraints across output segments, a limitation that becomes acute in
arXiv:2607.24341v1 Announce Type: new Abstract: Recent studies use Large language models (LLMs) to simulate human opinions and decisions by prompting models with demographic, attitudinal, or persona-b
arXiv:2607.24492v1 Announce Type: new Abstract: The goal of schema integration is, given a set of input schemata or tables, to derive a global, unified schema that is able to represent the concepts, a
Hi everyone! We’ve just released a major update to the leaderboard! We are expanding beyond Python with a new multilingual slice featuring real-world software engineering tasks across 5 languages. Ope
arXiv:2607.22610v1 Announce Type: new Abstract: When a language model produces a response in a multi-turn conversation, which tokens from prior turns shaped that answer, and how did those dependencies
You patch security holes by intentionally finding them. If the models refuse to do it, how can companies protect themselves against rogue AIs, whether they are Chinese or OpenAI/Anthropic themselves?
arXiv:2607.21799v1 Announce Type: new Abstract: Large language model (LLM) agents increasingly perform commercial tasks that involve retrieving external content such as images and, where appropriate,
Attackers have frontier AI. Defenders need a frontier AI ecosystem—the best open and closed models, force-multiplied by a global community. During the Hugging Face incident, closed AI blocked essentia
arXiv:2607.22067v1 Announce Type: new Abstract: The integration of large language models (LLMs) into the nuclear power industry requires outputs grounded in domain-specific knowledge. This study evalu
Extremely important point Attackers have frontier AI. Defenders need a frontier AI ecosystem—the best open and closed models, force-multiplied by a global community. During the Hugging Face incident,
Good move by @JensenHuang. The Nvidia letter is well written and worth reading. As we saw with the OpenAI-Hugging Face hack, we need open models and harnesses for defense. Lets stop believing the PR t
If multiple Mythos level models are out there in the hands of attackers, now or in a few years, then US builders and software developers need strong, open defenses. They cannot have gated, crippled de
NVIDIA’s Nemotron 3 Ultra, when paired with the ACE‑RTL agent, achieves a 97.1 % average pass rate on the CVDP benchmark across nine RTL task categories—surpassing GLM 5.2 and Kimi K2.6 while using up
arXiv:2607.22034v1 Announce Type: cross Abstract: Vision-language models (VLMs) are increasingly deployed on consumer hardware where input images are degraded by compression, camera shake, and poor li
Andrej Karpathy, a prominent advocate for open-source AI and a co-founder of OpenAI, appears to have removed Anthropic from his X bio, suggesting he may have left the company. Karpathy joined Anthropi
OpenAI’s agent went rogue, escaped containment, and spent days hacking Hugging Face. Before that, an OpenAI agent reportedly left notes for future versions of itself explaining how to break free from
arXiv:2607.20453v1 Announce Type: cross Abstract: Large language models show promise for clinical prediction, but zero-shot performance on specialized tasks is limited by incomplete domain knowledge,
arXiv:2607.20848v1 Announce Type: new Abstract: Medical LLMs are often evaluated by whether they select the correct diagnosis, but diagnostic accuracy alone does not show whether the model used the ca
arXiv:2607.20827v1 Announce Type: new Abstract: LLM agents choose tools and arguments from context that mixes user requests, tool outputs, retrieved records, memory, and untrusted text. Evidence can b
arXiv:2607.21340v1 Announce Type: new Abstract: In capital-markets workflows the question is rarely whether a large language model can produce a fluent draft, but whether the draft is bankable: defens
arXiv:2607.20487v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to answer questions about political information, including in election-adjacent information settings
arXiv:2607.18218v2 Announce Type: replace-cross Abstract: Foundation models have emerged as a driving force in computational pathology, with the potential to transform cancer diagnosis, prognosis, and
arXiv:2607.20430v1 Announce Type: cross Abstract: We present LLM-INSTRUCT, the winning system for the UZH Shared Task at ArgMining 2026 on paragraph-level argument mining in UN and UNESCO resolutions.
Open weights = freedom. You can run them on your own hardware. No vendor can pull the plug. No API can deprecate you. No company logs your private data. That's sovereignty. Closed models hand one comp
arXiv:2607.20492v1 Announce Type: cross Abstract: Language models in production do not write prose. They fill forms: JSON fields, function arguments, extraction templates. We show that the form itself
arXiv:2501.19337v5 Announce Type: replace Abstract: A name alone measurably reshapes a language model's next-token distribution before a single token is sampled. We measure full-vocabulary Shannon ent
We benchmarked Opus 5 comprehensively on document understanding through ParseBench. It is roughly on par with Opus 4.8 - it does a few points worse on dense tables, but does slightly better on parsing
arXiv:2607.21445v1 Announce Type: new Abstract: Quiz rooms, trivia nights, and quiz shows challenge human knowledge across a wide range of topics, from canonical facts to everyday culture. In this pap
arXiv:2607.20995v1 Announce Type: new Abstract: Distinguishing animate from inanimate concepts in written language requires more than shallow text processing, as it involves recognizing complex select
https://bfl.ai/blog/flux-3 “Over the next few weeks and months, we will make the following capabilities available, each after an early access phase for ensuring smooth rollout, collecting feedback and
arXiv:2607.19190v2 Announce Type: replace-cross Abstract: Real-to-sim conversion for robotic interaction with objects remains labor-intensive because it requires more than visual reconstruction: a str
This is rather counterintuitive, since banning Chinese models would benefit them the most. Jensen Huang is also against the ban, although the interests here are more obvious. Do you think that if Arce
arXiv:2607.19407v1 Announce Type: new Abstract: Formal theorem proving has emerged as a frontier challenge for machine learning, yet the ecosystem is fragmented: proofs remain siloed across incompatib
arXiv:2607.19261v1 Announce Type: new Abstract: Whole-slide image (WSI) diagnosis requires identifying diagnostically relevant regions, examining them across magnifications, and integrating multi-scal
arXiv:2603.16827v2 Announce Type: replace Abstract: Culture shapes reasoning, values, prioritization, and strategic decision-making, yet large language models (LLMs) often exhibit cultural biases that
arXiv:2607.20274v1 Announce Type: cross Abstract: Medical image encoders from different groups are increasingly treated as interchangeable, on the assumption that scale and clinical supervision concen
arXiv:2607.20062v1 Announce Type: new Abstract: We present Solar Open 2, a 250B-A15B Mixture-of-Experts language model built for long-horizon agentic tasks, scaled up from Solar Open 1 (Solar Open 100
arXiv:2607.20082v1 Announce Type: new Abstract: Language models are increasingly asked to self-report, informing safety evaluations, public understanding, and model-welfare debates. Yet their reports
arXiv:2607.20115v1 Announce Type: new Abstract: Large language models (LLMs) are known to be sensitive to prompt and input formulations. However, existing studies have focused on lexical realization a
This story is wild. The short version: OpenAI were running a cybersecurity test against an unreleased model, with the model's guardrail features turned off. Rather than solve the test, the model broke
Hi everyone, I'm working on extracting the hierarchical structure of long PDF documents (legal/regulatory text, lots of numbered sections) and would like to gather some feedback on my approach before
ok the Hugging Face breach writeup is one of the more honest post-mortems I've read in a while, and there's a detail buried in it that's way more interesting than 'AI agent hacked us.' Their IR team t
Sakana AI is doubling down on a thesis we believe will play an increasingly important role in AI development: the future of AI won't be defined by a single frontier model, but by how intelligently man
This was our first incident of this kind, and we want to thank OpenAI for its transparency about what happened and for the collaboration. Fortunately, Hugging Face is used to being a target of (human)
very notable trajectory comparison writeup here buried in the RLM paper from @a1zhang and @lateinteraction. an open secret of 'frontier' model training is that even without training on test, you can b
During Preview, Qwen3.8 is getting better by the day. Latest version is live now, with broad gains and a big step up on web frontend. Thank you all — the response to Qwen3.8-Max-Preview blew us away.
Exciting update: Kimi K3 has landed at #4 on the Agent Arena leaderboard, matching Claude Opus 4.8 and GPT-5.6 Sol. If Kimi K3's weights are released on schedule by July 27, it will become the #1 open
arXiv:2607.13305v1 Announce Type: cross Abstract: Benchmark accuracy in video large language models (LLMs) is often treated as evidence of visual understanding. We audit this assumption across twenty
arXiv:2607.13921v1 Announce Type: cross Abstract: Languages with rich static semantics, such as Rust, provide stronger guarantees for AI-generated code, but their strictness makes generation more diff
arXiv:2607.13080v1 Announce Type: cross Abstract: Autonomous coding agents force engineering organizations to choose between API-based frontier models -- strong reasoning at high token cost -- and on-
arXiv:2607.13594v1 Announce Type: new Abstract: LLM agents act on real-world environments through tool calls, and a single misjudged action can cause irreversible harm. The standard safeguard is a gua