Knowledge catalogue
arXiv:2607.22716v1 Announce Type: new Abstract: In this paper, we show for the first time that visual token pruning enhances the robustness of Multimodal Large Language Models (MLLMs), mitigating vuln
arXiv:2607.22868v1 Announce Type: new Abstract: Runtime guardrails act before irreversible tool calls, but their guarantees depend on what policy state is representable, what a judge observes, and whe
arXiv:2607.23271v1 Announce Type: cross Abstract: Contrastive vision-language models such as CLIP map semantically opposite phrases (e.g., 'a dog' vs. 'not a dog') to nearly identical embeddings, rend
arXiv:2607.23469v1 Announce Type: cross Abstract: Photonic-crystal surface-emitting lasers (PCSELs) can combine high-power operation with narrow-divergence surface emission, but optimizing coupled par
arXiv:2607.24010v1 Announce Type: new Abstract: Active RAG systems decide when to retrieve external knowledge during generation, making them a budget-sensitive case of agentic RAG and self-adaptive re
arXiv:2607.23624v1 Announce Type: cross Abstract: Third-party API routers have become a common layer that unifies access across increasingly diverse LLM providers. In coding-agent workflows, high-auto
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?
It doesn't solve any problems. Just a few paragraphs above, he says he fears that authoritarian states (he names China, and possibly others) can use their models to do evil stuff. And surely enough, m
Why is this breaking news? And what the heck does it actually mean? Has anyone actually defined it? Is it anything more than a ruse to distract from a disturbing hack and falling confidence in AI? BRE
arXiv:2607.22595v1 Announce Type: new Abstract: Mechanistic interpretability (MI) has emerged as a powerful approach for analyzing and intervening in inference computations, with a growing number of a
arXiv:2607.22656v1 Announce Type: cross Abstract: Objective: This paper investigates the gendered structure of speaker addressee relationships in film dialogue, asking not merely who speaks, but who i
arXiv:2607.21632v1 Announce Type: new Abstract: Traditional benchmarks for LLMs primarily rely on static datasets and objective scoring metrics, which often fail to capture differences in response qua
arXiv:2607.21960v1 Announce Type: new Abstract: This paper demonstrates how deep reinforcement learning (DRL) enables adaptive locomotion of snake-like robots in dynamically changing viscous environme
arXiv:2607.21725v1 Announce Type: new Abstract: General-purpose robots need to reason about their actions, combining perception, world knowledge, planning, success detection, recovery, and low-level c
arXiv:2607.21646v1 Announce Type: new Abstract: Ensuring safety in reinforcement learning under nonstationarity requires determining whether a learning system can safely adapt to forecasted environmen
arXiv:2607.21804v1 Announce Type: cross Abstract: Lossless acceleration schemes, such as speculative decoding, promise significant inference speedups by relying on dynamic token-level alignment betwee
arXiv:2607.21619v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) have achieved impressive performance, but their safety alignment remains vulnerable to jailbreak attacks. Exist
arXiv:2607.22241v1 Announce Type: new Abstract: Recent advances in video diffusion models have spurred interest in human-object interaction (HOI) video generation, which demands fine-grained control o
創業以来、オープンソースコミュニティから多くを学び、また研究成果の公開を通じてそこに貢献してきました。オープンなエコシステムが健全なAI産業と技術主権を支える重要な基盤の一つであると考えており、その発展を支持します。 このたび、Sakana AIは、オープンウェイトAIモデルに関する公開書簡 「Open Weights and American AI Leadership」に署名しました。 書簡は
arXiv:2601.17483v2 Announce Type: replace Abstract: Training modern neural networks is increasingly fragile, with rare but severe destabilizing updates often causing irreversible divergence or silent
arXiv:2607.22264v1 Announce Type: new Abstract: Autoregressive foundation models trained on tokenized electronic health records (EHRs) can support zero-shot clinical prediction, yet most operate on st
For the enterprise, the promise of agentic AI is much more than just a better chatbot. It is software agents that execute business tasks end-to-end across people, business workflows, data, and systems
arXiv:2607.21839v1 Announce Type: cross Abstract: Privacy-preserving machine learning auditing protocols allow auditors to assess models for properties such as accuracy or fairness, without revealing
arXiv:2607.21623v1 Announce Type: new Abstract: We present EaaS, a cloud-native reference architecture that operationalizes AI evaluation methods as six stateless Kubernetes microservices: conformal p
arXiv:2509.10058v2 Announce Type: replace Abstract: Accurate color alignment in text-to-image (T2I) generation is critical for applications such as fashion, product visualization, and interior design,
arXiv:2607.22078v1 Announce Type: new Abstract: As autonomous driving systems move toward real-world deployment, interpretable, behavior-level decision-making is essential for safety, trust, and regul
arXiv:2607.22062v1 Announce Type: cross Abstract: This paper addresses the modeling and synthesis of constrained robotic system behaviors using Petri nets (PNs). It investigates how to construct model
arXiv:2607.22012v1 Announce Type: new Abstract: Off-Policy Evaluation and Learning (OPE/L) in contextual bandits is rapidly gaining popularity in real systems because new policies can be evaluated and
arXiv:2601.12311v2 Announce Type: replace-cross Abstract: The emergence of 6G-enabled vehicular metaverses enables Autonomous Vehicles (AVs) to operate across physical and virtual spaces through space
arXiv:2607.22446v1 Announce Type: new Abstract: Recent radiance field methods represent scenes with 2D primitives that offer surface alignment and efficient rasterization, from Gaussian disks to trian
arXiv:2607.22249v1 Announce Type: new Abstract: Nociception is a protective biological mechanism that links harmful stimulation to a reaction. This paper investigates artificial nociception for a robo
arXiv:2607.21627v1 Announce Type: cross Abstract: End-to-end reinforcement learning can improve the accuracy of compound LLM systems, but it does not constrain how modules divide labor internally. We
arXiv:2606.13842v2 Announce Type: replace Abstract: We present an algorithm for efficient domain-adaptive policy learning via kernel representations. Learning domain-adaptive policies is challenging s
arXiv:2607.22020v1 Announce Type: new Abstract: Learned manipulation policies increasingly predict motions for abstract 'hands' and are attractive in practice because they rely on easily collected dem
arXiv:2506.18748v2 Announce Type: replace-cross Abstract: We consider resource allocation problems in multi-user wireless networks, where the goal is to optimize a network-wide utility function subjec
arXiv:2607.21608v1 Announce Type: cross Abstract: With the EU AI Act entering into force, organizations developing or operating AI systems face new obligations on transparency, risk management, and tr
arXiv:2607.21660v1 Announce Type: cross Abstract: Artificial intelligence (AI) is accelerating materials prediction and design by enabling efficient exploration of chemical and structural spaces, with
arXiv:2607.21661v1 Announce Type: new Abstract: Diffusion policies generate multimodal robot action sequences from demonstrations, but steering them toward deployment-time constraints typically relies
Great technical paper from Harvard and MIT. It's on role drift in compound LLM systems. (bookmark it) End-to-end RL improves the accuracy of a multi-module LLM pipeline without constraining how the mo
arXiv:2607.22299v1 Announce Type: cross Abstract: Forecasting multiple time-series with high-dimensional covariates presents a core challenge: unifying common temporal patterns while retaining meaning
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
Open source software is a critical pillar of the global economy. It underpins cloud computing, financial services, manufacturing, telecommunications, government and internet services by making technol
arXiv:2607.22200v1 Announce Type: new Abstract: End-to-end OCR systems based on vision-language models have achieved strong performance in complex document OCR, but their efficiency is limited by the
arXiv:2607.22166v1 Announce Type: new Abstract: Path planning under partial observability remains challenging because an agent must make long-horizon navigation decisions from only locally bounded obs
arXiv:2607.10023v2 Announce Type: replace-cross Abstract: Understanding music requires understanding localized relationships across data modalities, e.g., how time in performance audio maps onto posit
arXiv:2607.15456v2 Announce Type: replace-cross Abstract: Looped, weight-tied Transformers reduce parameters by reusing a single block, but decoding still stores a separate K/V cache for every recurre
arXiv:2607.21949v1 Announce Type: new Abstract: Continual temporal knowledge graph (TKG) reasoning aims to continuously incorporate newly emerging facts while preserving previously acquired knowledge.
arXiv:2603.28956v2 Announce Type: replace-cross Abstract: The minimum-norm interpolator (MNI) framework has recently attracted considerable attention as a tool for understanding generalization in over
arXiv:2607.21653v1 Announce Type: cross Abstract: Agentic reinforcement learning research is constant algorithm modification, new estimators, new pipeline stages, new rollout schemes, and in mainstrea
arXiv:2607.22083v1 Announce Type: cross Abstract: We present Nanbeige4.2-3B, a compact general agentic model with 3B non-embedding parameters. It delivers strong performance across code-agent, office-
New research from NVIDIA. They just dropped a PyTorch-native training framework for agentic RL. (bookmark it) Paper summary: Molt is a PyTorch-native agentic RL framework with an unusual design target
Jaspreet Singh / Reuters: Nvidia forms the Open Secure AI Alliance, a coalition including CrowdStrike, Hugging Face, and Dell to develop and share tools for AI safety and cybersecurity — Nvidia (NVDA.
Nvidia on Monday said it is joining forces with Microsoft, SpaceX, IBM, and other tech companies to build and share open-source AI security tools. The new Open Secure AI Alliance said open tools are r
arXiv:2607.22430v1 Announce Type: new Abstract: Learning world models that infer environment dynamics from high-dimensional observations and predict outcomes under candidate actions is central to plan
arXiv:2501.18530v3 Announce Type: replace-cross Abstract: We consider a teacher-student model of supervised learning with a fully-trained two-layer neural network whose width k and input dimension d a
arXiv:2607.21670v1 Announce Type: cross Abstract: Action tokenization maps continuous robot action chunks to discrete tokens and has become an important interface for modern visuomotor policies. Exist
arXiv:2607.22518v1 Announce Type: cross Abstract: In this paper, we propose a new solution for addressing the content cold-start problem in industry-scale search and recommender systems. Compared to p
arXiv:2607.22114v1 Announce Type: new Abstract: Autoregressive foundation models for electronic health records (EHRs) typically inherit pretraining methods from language modeling, where patient trajec
arXiv:2607.21774v1 Announce Type: new Abstract: Large language models may infer demographic attributes from subtle linguistic cues even when those attributes are not explicitly stated. This pilot stud
arXiv:2607.22117v1 Announce Type: new Abstract: Domain Generalization (DG) aims to learn representations robust to distribution shifts. Recent geometric alignment methods, such as CPCANet, extract dom