Knowledge catalogue
OpenAI continues to be the most disconcerting of the major AI companies. In 2023, OpenAI pledged 20% of its entire compute to AI safety. Last month it quietly dissolved the team whose only job was cat
arXiv:2608.14522v1 Announce Type: new Abstract: As AI systems make more morally loaded decisions across society, one response has been moral preference elicitation. In this approach, researchers poll
arXiv:2608.14082v1 Announce Type: new Abstract: Autonomous navigation in cluttered environments is hampered by partial observability and dynamic constraints. This paper presents PILOT, a constraint-aw
arXiv:2608.14142v1 Announce Type: new Abstract: Open-vocabulary object detection test-time adaptation (OVOD-TTA) aims to address the performance degradation that pre-trained base models suffer when en
arXiv:2608.13969v1 Announce Type: new Abstract: Prompt tuning adapts CLIP-based vision-language models with few trainable parameters, yet its predictions remain sensitive to the spatial sampling impos
arXiv:2608.13671v1 Announce Type: new Abstract: Modern text-to-image models can generate highly realistic images from natural-language prompts, while recent advances in prompt inversion have made it i
arXiv:2608.13563v1 Announce Type: cross Abstract: Early-stage teams often lack users, time, and budget to run repeated UX studies, yet still need decision-oriented signals to iterate safely. We study
arXiv:2608.14319v1 Announce Type: new Abstract: We study quantum multi-armed bandits (QMAB) and quantum linear bandits (QLB) in the model of Wan et al. [2023], where the learner queries each arm or ac
arXiv:2506.14488v2 Announce Type: replace-cross Abstract: Structure-based drug design (SBDD) models are central to modern pharmaceutical research, enabling the rational exploration of protein-ligand i
arXiv:2507.02169v2 Announce Type: replace Abstract: Machine learning models are often used in applications where their inputs change due to routine interactions, strategic manipulation, or noise. In s
arXiv:2510.02916v2 Announce Type: replace-cross Abstract: We propose SALSA-V, a multimodal video-to-audio generation model capable of synthesizing highly synchronized, high-fidelity long-form audio fr
arXiv:2608.14286v1 Announce Type: cross Abstract: Vision language models (VLMs) are increasingly used in industrial decision-making systems, such as recruitment support and recommendation. This motiva
arXiv:2602.06746v2 Announce Type: replace Abstract: We study multi-task reinforcement learning (RL), a setting in which an agent learns a single, universal policy capable of generalising to arbitrary,
arXiv:2608.14306v1 Announce Type: new Abstract: This paper presents a coordination architecture for heterogeneous UAV/UGV swarms that synthesises mission actions from uncertain, multi-modal sensor evi
Some say dogs look like their owners. Generative AI CEOs have come to sound like their product: frequently wrong, never in doubt, often full of BS. Dario is full of sh@t. Birds of a feather. JUST SAY
Some thoughts on Dario’s post: 1. Dario does not actually address Gavin Baker’s account of what he said – something he could easily deny if it were inaccurate. 2. Dario claims his critics live in a “b
arXiv:2608.14531v1 Announce Type: new Abstract: Articulated vehicles are the workhorses of freight transportation, and their autonomous navigation is challenging. Their physical characteristics and mo
arXiv:2608.14024v1 Announce Type: new Abstract: Vision-language-action (VLA) models for autonomous driving jointly produce scene interpretation, language-based reasoning, and driving trajectories. Exi
arXiv:2608.13800v1 Announce Type: new Abstract: Transition path sampling (TPS) aims to efficiently generate rare molecular transition trajectories between metastable states and is essential for unders
arXiv:2608.14177v1 Announce Type: cross Abstract: Deep spatiotemporal models integrating graph convolutions and attention mechanisms have demonstrated excellent performance in network-level traffic fl
arXiv:2504.21307v3 Announce Type: replace Abstract: Diffusion models excel at generating high-quality images but can memorize and reproduce harmful concepts when prompted. Although fine-tuning methods
arXiv:2505.16882v3 Announce Type: replace Abstract: Ethological research increasingly benefits from the growing affordability and accessibility of drones, which enable the capture of high-resolution f
arXiv:2608.14211v1 Announce Type: cross Abstract: Tabular Foundation Models (TFMs) have emerged as leading methods for tabular predictive tasks, leveraging in-context learning to predict on new data w
arXiv:2608.13596v1 Announce Type: cross Abstract: Heterogeneous model fusion seeks to combine models that differ in tasks, initializations, architectures, or scales. We study an underexplored cross-sc
arXiv:2608.14392v1 Announce Type: new Abstract: Neuron- and path-level interventions offer the finest-grained route to defending large language models (LLMs) against jailbreak attacks, yet existing me
arXiv:2511.19009v2 Announce Type: replace-cross Abstract: Large language models (LLMs) demonstrate powerful capabilities across various natural language processing tasks,yet their inherent safety vuln
arXiv:2608.13613v1 Announce Type: cross Abstract: Recent breakthroughs in generative models have made text-to-voice generation (TTV) possible, enabling the synthesis of speech directly from textual vo
arXiv:1912.08786v3 Announce Type: replace-cross Abstract: Three generations of software have transformed the role of artificial intelligence in society. In the first, programmers wrote explicit logic.
arXiv:2608.13861v1 Announce Type: new Abstract: Morphing attacks pose a serious threat to face recognition systems. However, existing image-based morphing attack detection (MAD) methods often generali
OpenAI has shut down the internal team that evaluated whether its AI models could pose “catastrophic risks,” a decision reported by the Financial Times on August 16, 2026. The move was announced via a
This is The Stepback, a weekly newsletter breaking down one essential story from the tech world. For more on AI safety, follow Robert Hart. The Stepback arrives in our subscribers' inboxes at 8AM ET.
@sethharpesq OpenAI spent $1.60 for every dollar it earned in 2025. We need more journalism that cuts through all of the accounting gimmicks to get to the heart of what is really happening in the US A
Dr. Catharine Young noted on Twitter that someone should pursue an internship at the U.S. Food & Drug Administration (FDA). On August 16, Polymarket reported that Anthropic CEO Dario Amodei predicted
1/2 Thanks Gavin for an especially thoughtful exchange. I don't usually spend much time on social media but I wanted to engage here because it really brings out the heart of an important conversation.
2/2 Second, on the messaging around AI. I do not agree that my messaging has been disproportionately negative. In fact it has been about equally balanced between risks and benefits: I’ve written one m
Financial Times: Sources: OpenAI's repeated exec reshuffles and departures have frustrated some staff as it prepares for an IPO; it disbanded its “preparedness” team in July — Executive exits and safe
The FT reports this weekend that OpenAI quietly disbanded its preparedness team at the end of last month, with responsibilities for individual areas of the preparedness framework distributed across ot
Wrong! If an AI leverages symbolic operations (conditionals, operations over variables, code interpreters, etc) and neural networks it is neurosymbolic. If it doesn’t, it’s not. ( I laid all of this o
$3 trillion in capex into a technology that has zero correlation with value creation what could possibly go wrong? wow! new OpenAI study shows “no correlation between how often employees use AI and ho
arXiv:2608.13511v1 Announce Type: new Abstract: Repeated in-situ evaluation of ocean-glider planners requires scarce vehicles, operators, deployment and recovery resources, and ocean conditions that c
arXiv:2608.12869v1 Announce Type: cross Abstract: Probabilistic Circuits (PCs) are generative models that support exact inference and, unlike deep neural networks, admit an exact and tractable measure
arXiv:2608.13467v1 Announce Type: new Abstract: We study the Moreau--Yosida unadjusted Langevin algorithm (MYULA) for the nonsmooth composite target [ pi(dx)propto exp{-f(x)-g(x)},dx, qquad xinmathbb
arXiv:2608.12424v1 Announce Type: cross Abstract: This study focuses on developing an AI-supported prototype for multiperspective interest rate forecasting that combines classical econometric models w
arXiv:2602.11143v3 Announce Type: replace Abstract: Humanoid locomotion has advanced rapidly with deep reinforcement learning (DRL), enabling robust feet-based traversal over uneven terrain. Yet platf
arXiv:2608.12788v1 Announce Type: new Abstract: The rapid advancement of Auto-Research has surfaced a fundamental evaluation challenge: how can we measure the alignment, logical coherence, and evoluti
arXiv:2608.13422v1 Announce Type: new Abstract: Visual bottlenecks that focus policy inputs on regions of interest (ROIs) can improve data-efficient visuomotor learning by separating where to look fro
arXiv:2511.22651v2 Announce Type: replace-cross Abstract: Optimization methods have long advanced many fields, yet they struggle when faced with design problems where the search space and design param
arXiv:2608.13160v1 Announce Type: cross Abstract: Multilingual retrieval-augmented generation (mRAG) equips large language models with access to globally distributed external knowledge for complex mul
arXiv:2608.12764v1 Announce Type: cross Abstract: Deep search agents operate over trajectories spanning dozens of steps, yet standard reinforcement learning provides only a single outcome reward per t
arXiv:2608.12971v1 Announce Type: new Abstract: Face recognition systems have been shown to be biased toward certain demographic groups by exhibiting different error rates across gender, age, or ethni
arXiv:2608.13118v1 Announce Type: new Abstract: Verification of neural networks against relational specifications, such as global robustness, is crucial for safety-critical applications of cyber-physi
arXiv:2608.13209v1 Announce Type: cross Abstract: Many operational decisions are sequences of interventions under a cumulative resource limit, such as a maintenance schedule within a crew-hour budget.
arXiv:2608.13438v1 Announce Type: cross Abstract: Contact-rich manipulation failures are often detected only after the robot has committed to contact. This is especially limiting in wrist-camera setup
arXiv:2608.13391v1 Announce Type: new Abstract: Interactive autoregressive video generation demands both low-latency rollouts and precise online control. Few-step distillation accelerates generation b
arXiv:2608.13387v1 Announce Type: new Abstract: On-policy distillation (OPD) supervises a student language model on trajectories sampled from its current policy, but assigns equal credit to response t
arXiv:2608.12773v1 Announce Type: new Abstract: Semi-supervised semantic segmentation has long turned on one question, which pseudo-labels to trust, and a generation of selection rules, dynamic thresh
arXiv:2608.01735v2 Announce Type: replace Abstract: On-policy (self) distillation (OPSD) is increasingly adopted for language-model post-training. It strengthens the teacher with privileged informatio
arXiv:2608.13474v1 Announce Type: new Abstract: Vision-language-action models (VLAs) are moving rapidly towards deployment as general-purpose manipulation policies, but we currently lack basic tools f
arXiv:2608.12913v1 Announce Type: new Abstract: Contrastive Decoding (CD) improves generation quality, but its amateur-model pass makes decoding expensive. Accelerating CD with speculative decoding ra
arXiv:2510.21805v2 Announce Type: replace-cross Abstract: Generative recommendation (GR) is an emerging paradigm that represents each item via a tokenizer as an n-digit semantic ID (SID) and predicts