safety

arXiv:2509.21172v2 Announce Type: replace Abstract: Inverse reinforcement learning (IRL) aims to infer rewards from observed behavior, but rewards are not identified from the policy alone: many reward

arXiv:2510.01569v3 Announce Type: replace Abstract: We present InvThink, a training and prompting framework that requires the model to enumerate, analyze, and constrain potential failures before gener

arXiv:2605.07514v1 Announce Type: cross Abstract: World Action Models (WAMs) enable decision-making through imagined rollouts by predicting future observations and actions. However, the reliability of

It is insane that we allow big tech to race toward superintelligence without any oversight. This is a grave national security risk. Industry will not regulate itself. Get trained with Torchbearer to a

arXiv:2605.07865v1 Announce Type: cross Abstract: On-Policy Distillation (OPD) has emerged as a dominant post-training paradigm for large language models, especially for reasoning domains. However, OP

arXiv:2605.06934v1 Announce Type: new Abstract: We augment the Slotine--Li adaptive controller for Euler--Lagrange systems with three learned components: a structured-quadratic Lyapunov function (V_ps

arXiv:2605.07026v1 Announce Type: cross Abstract: Functional connectivity (FC) derived from resting-state fMRI is widely used to characterize large-scale brain network alterations in neurological and

arXiv:2605.07064v1 Announce Type: new Abstract: How to achieve vision-language (VL) tracking using natural language descriptions from a video sequence extbf{without relying on any bounding-box ground

arXiv:2605.07079v1 Announce Type: cross Abstract: World models predict future transitions from observations and actions. Existing works predominantly focus on image generation only. Visual feature-bas

arXiv:2605.07495v1 Announce Type: new Abstract: Unpaired smartphone ISP is a challenging problem due to the lack of scene and color alignment between RAW and target RGB images. Many existing methods e

arXiv:2605.06933v1 Announce Type: new Abstract: Our computing ecosystem is being transformed by two emerging paradigms: the increased deployment of agentic AI systems and advancements in quantum compu

arXiv:2605.07835v1 Announce Type: new Abstract: Multi-robot systems in automated warehouses must manage continuous streams of pickup-and-delivery tasks while ensuring efficiency and safety. Prior work

arXiv:2507.21183v5 Announce Type: replace-cross Abstract: As the era of large language models (LLMs) unfolds, Preference Optimization (PO) methods have become a central approach to aligning LLMs with

arXiv:2605.07178v1 Announce Type: new Abstract: Remote sensing change detection is pivotal for urban monitoring, disaster assessment, and environmental resource management. Yet, unimodal deep learning

Michael Burry urged investors to scale back exposure to surging technology stocks, saying the current market environment has reached historically dangerous extremes reminiscent of prior speculative bu

arXiv:2605.07550v1 Announce Type: new Abstract: 3D vision systems are fundamentally constrained by their reliance on visual overlap: reconstruction methods require it for geometric alignment, while ge

arXiv:2601.04731v2 Announce Type: replace Abstract: Current critic-free RL methods for large reasoning models suffer from severe inefficiency when training on positive homogeneous prompts (where all r

arXiv:2605.08050v1 Announce Type: new Abstract: Talking-head generation requires joint modeling of identity, head pose, facial expression, and mouth dynamics. Existing methods typically address only a

arXiv:2602.07026v2 Announce Type: replace-cross Abstract: Despite the success of multimodal contrastive learning in aligning visual and linguistic representations, a persistent geometric anomaly, the

arXiv:2605.07370v1 Announce Type: cross Abstract: V2X can warn an autonomous vehicle about hazards beyond line-of-sight, but it also brings uncertainty: messages may be delayed, dropped, or even forge

arXiv:2605.08024v1 Announce Type: new Abstract: Learning-to-defer (L2D) can make glaucoma screening safer by routing difficult/uncertain cases to humans, yet standard formulations overlook expert avai

arXiv:2601.07247v2 Announce Type: replace-cross Abstract: Learning models that can handle distribution shifts is a key challenge in domain generalization. Invariance learning, an approach that focuses

arXiv:2605.07537v1 Announce Type: new Abstract: Partially Observable Markov Decision Processes (POMDPs) are systems in which one agent interacts with a stochastic environment, and receives only partia

arXiv:2603.16876v2 Announce Type: replace-cross Abstract: We propose MARL-Rad, a multi-modal multi-agent reinforcement learning framework for radiology report generation that trains the entire agentic

arXiv:2605.06864v1 Announce Type: new Abstract: We study multi-objective multi-agent multi-armed bandits (MO-MA-MAB) under stochastic rewards, where agents observe heterogeneous reward vectors and com

Mythos identified one genuine vulnerability in cURL while also reporting three false positives and one issue classified as a bug during their security analysis. The post references cURL's founder or l

@NameInteger @GaryMarcus @geoffreyhinton Hinton's argument was about encoding: LLMs don't encode text as text, but as weighted matrices. It makes no difference. Memorised data is memorised data no mat

arXiv:2503.04638v3 Announce Type: replace Abstract: Most Continual Learning (CL) methods maintain performance on earlier tasks by storing exemplars in a replay buffer, introducing memory overhead that

arXiv:2605.07794v1 Announce Type: new Abstract: World Action Models (WAMs) are an emerging family of policies that tie robot action generation to future-observation modeling. In this work, we focus on

Not even surprised by horrific stories like these anymore. The mission of getting LLMs aligned with human values has largely been a failure. NEW: ChatGPT advised the FSU shooter that a mass shooting w

arXiv:2604.03675v2 Announce Type: replace Abstract: Agentic search enables language models to solve knowledge-intensive tasks by adaptively acquiring external evidence over multiple steps. Reinforceme

arXiv:2605.08031v1 Announce Type: new Abstract: Vision-language models (VLMs) raise growing concerns about privacy, copyright, and bias, motivating machine unlearning to remove sensitive knowledge. Ho

arXiv:2605.07393v1 Announce Type: new Abstract: A fundamental challenge in model-based offline reinforcement learning (RL) lies in the trade-off between generalization and robustness against exploitat

oh. my. god. 😱 FT Exclusive: NHS England has granted external staff from companies including Palantir “unlimited access” to identifiable patient data while working on a part of its flagship data platf

arXiv:2602.08786v4 Announce Type: replace-cross Abstract: There is an extensive literature that studies how to find optimal policies in resource allocation problems, taking the underlying design param

arXiv:2605.06732v1 Announce Type: new Abstract: State-of-the-art model-based reinforcement learning methods train policies on imagined rollouts. These rollouts are trajectories generated by a learned

One estimate of how much annual revenue AI needs to “make sense”: 1.6 trillion. That’s four times what Google made in its best year. (total revenue so far is perhaps on order of 100 billion.) In 2024

arXiv:2605.07931v1 Announce Type: cross Abstract: Vision-language-action (VLA) models increasingly rely on auxiliary world modules to plan over long horizons, yet how such modules should be parameteri

arXiv:2605.07080v1 Announce Type: new Abstract: Many real-world resource allocation systems, such as humanitarian logistics and vaccine distribution, must preposition limited supply across multiple lo

Openclaw token consumption dropped 50% within a month according to OpenRouter usage data, as reported by Gary Marcus. The post likely discusses potential causes for this significant decline, such as c

arXiv:2605.07649v1 Announce Type: cross Abstract: Over the last few years, research on autonomous systems has matured to such a degree that the field is increasingly well-positioned to translate resea

arXiv:2605.07598v1 Announce Type: new Abstract: Actionable Recourse provides individuals with actions they can take to change an unfavorable classifier outcome. While useful at the instance level, it

arXiv:2605.07414v1 Announce Type: cross Abstract: Tool-calling text-to-image (T2I) agents can plan and execute multi-step tool chains to accomplish complex generation and editing queries. However, thi

arXiv:2605.07039v1 Announce Type: new Abstract: Large language models have become drivers of evolutionary search, but most systems rely on a fixed, prompt-elicited policy to sample next candidates. Th

arXiv:2605.07055v1 Announce Type: cross Abstract: Foundation models (FMs) have shown great promise in medical imaging, but most FMs are trained on unimodal data within isolated domains, such as brain

arXiv:2605.07248v1 Announce Type: new Abstract: Beyond training-time optimization, scaling test-time computation has emerged as a key paradigm to extend the reasoning capabilities of Large Language Mo