Knowledge catalogue
Over the last few years, we have made significant progress on prompt injection for Claude models. As agents do more, it is increasingly urgent that we solve it across the industry. Really excited to p
arXiv:2411.06403v4 Announce Type: replace Abstract: Why can neural networks struggle to learn optimal play in a game with a simple, efficient solution algorithm? Motivated by empirical difficulties wi
I remember great creativity from GPT-2. It was trained on a dataset that is orders smaller than current training datasets. The problem with overfitting a model to the training data is that it becomes
arXiv:2610.16543v1 Announce Type: new Abstract: Multi-agent discussion produces reasoning collaboratively, while deployment still relies on a smaller student that must answer alone. Full-graph distill
arXiv:2610.19601v1 Announce Type: new Abstract: Knowledge distillation (KD) improves language-model pretraining by matching teacher distributions at corpus prefixes, but provides no direct supervision
arXiv:2610.19248v1 Announce Type: new Abstract: Robotic disassembly of end-of-life (EoL) products requires robots to accommodate product variation, follow prerequisite sequences, and approach componen
arXiv:2610.18784v2 Announce Type: new Abstract: Long-horizon agents accumulate progress in persistent task state, yet repeatedly process interaction histories describing how that progress was achieved
arXiv:2610.19123v1 Announce Type: new Abstract: Generating novel molecules with desired properties is a central challenge in drug discovery. Although foundational work focused on de novo generation, p
arXiv:2610.15208v1 Announce Type: cross Abstract: Deep research requires connecting multiple steps of search, reasoning, and computation, with intermediate findings guiding what to investigate next. A
arXiv:2602.11965v2 Announce Type: replace Abstract: Temporal domain generalization (TDG) learns from a sequence of observed domains to predict models for unseen future domains. Existing parameter-dyna
arXiv:2610.19438v1 Announce Type: new Abstract: Vision-language-action (VLA) policies fine-tuned on LIBERO are trained and evaluated on the same ten instruction strings, and rewording the instruction
arXiv:2610.15797v1 Announce Type: new Abstract: Planning and revision are intended to bring generated code closer to the task specification. Yet a mistaken interpretation can persist across the plan,
arXiv:2610.17059v1 Announce Type: new Abstract: Clinical decisions depend on anticipating how a patient's condition will evolve as treatment is adjusted and new evidence becomes available. Existing me
arXiv:2610.16466v1 Announce Type: cross Abstract: We present PatMT2 (Patent Machine Translation, v2), a two-stage LoRA pipeline that adapts the Hy-MT2-7B dedicated translation model to four-language (
arXiv:2610.16538v1 Announce Type: cross Abstract: We present PatTransBench, a 12,000-item Chinese-English/Japanese/Korean patent machine translation (PatMT) benchmark with a four-dimensional evaluatio
arXiv:2610.20240v1 Announce Type: new Abstract: A multiplayer world model must predict how the independent actions of several players change a shared environment, including changes outside a player's
arXiv:2610.14855v1 Announce Type: new Abstract: The growing deployment of compact large language models on smartphones is enabling autonomous on-device decision-making, such as emerging mobile GUI age
arXiv:2610.14636v1 Announce Type: new Abstract: Terminal rewards make multi-turn agents trainable without step-level annotations, but group-relative advantages provide little guidance about which inte
arXiv:2610.04693v2 Announce Type: replace Abstract: Agents are entering finance, healthcare and law, sectors where a violation leaves no lexical signature and carries real penalties. Whether an omissi
Reed Albergotti / Semafor: Perplexity open sources SPACE, its AI agent sandbox, and says partners including CoreWeave, Akamai, and Crusoe agreed to offer SPACE to customers — THE SCOOP — Perplexity is
arXiv:2610.19314v1 Announce Type: new Abstract: Large language model (LLM) evaluators are increasingly used to assess personalized assistants, yet whether they can reliably judge personalization has r
arXiv:2610.20548v1 Announce Type: new Abstract: Understanding the social intelligence of large language models (LLMs) is increasingly important for applications involving human-AI interaction and mult
arXiv:2610.16141v1 Announce Type: new Abstract: LLM agents increasingly operate over multiple entities, where reliable response generation requires not only retrieving relevant evidence but also prese
arXiv:2610.19823v1 Announce Type: new Abstract: Text-to-video (T2V) generation models can produce visually realistic videos that nevertheless violate the physical consequences implied by the prompt an
arXiv:2610.17288v1 Announce Type: new Abstract: Building physical simulations from video requires coding agents to translate observations into executable models and refine them through tool use. We in
arXiv:2610.20560v1 Announce Type: cross Abstract: 3D object composition is pivotal to digital content creation and robot simulation, while also serving as a critical testbed for physical intelligence
arXiv:2610.18946v1 Announce Type: cross Abstract: Physical-layer spoofing attacks pose a critical threat to wireless localization systems by manipulating channel state information (CSI). Detecting suc
arXiv:2610.20524v1 Announce Type: new Abstract: Existing robot policy memories typically evolve with observation sequences, implicitly treating each update as the same amount of physical time. Yet ide
arXiv:2503.21821v2 Announce Type: replace Abstract: We introduce Physics, a benchmark for evaluating foundation models on advanced physics problem solving. It contains 1034 annotated, open-ended probl
arXiv:2610.18886v1 Announce Type: cross Abstract: Natural supercavitation behind conical cavitators generates a vapor-filled envelope that, when sustained over the full body of a high-speed underwater
arXiv:2610.19763v1 Announce Type: new Abstract: Physics-guided networks for underwater image restoration predict transmission and backscatter and invert the image-formation model, yet rarely test whet
arXiv:2506.20642v3 Announce Type: replace Abstract: Chain-of-Thought (CoT) prompting significantly enhances large language models' (LLMs) problem-solving capabilities, but still struggles with complex
arXiv:2610.20579v1 Announce Type: new Abstract: Unlike singles, humanoid table tennis doubles requires teammates to strike in strict alternation while sharing a confined workspace. Rapid repositioning
arXiv:2603.23678v2 Announce Type: replace Abstract: Large Language Models (LLMs) offer transformative solutions across many domains, but healthcare integration is hindered by strict data privacy const
Tool-augmented LLM agents discover relevant APIs from large registries before planning execution, yet finite context windows prohibit injecting all tool definitions into a single prompt. The dominant
arXiv:2606.22953v2 Announce Type: replace Abstract: Context managers summarize or evict old messages so agents can operate beyond a finite window. This creates a reliability problem: removing history
arXiv:2610.19131v1 Announce Type: cross Abstract: Maintenance work orders are the largest written record of how industrial equipment fails, yet in most plants they are close to unusable for analysis.
arXiv:2610.20584v1 Announce Type: new Abstract: Computer-use agents have demonstrated strong capabilities in interpreting screens and executing keyboard and mouse actions. However, games, with their v
arXiv:2610.19011v1 Announce Type: new Abstract: The Practical Molecular Optimization (PMO) benchmark standardized evaluation in molecular optimization, but it is built on simple property-based oracles
arXiv:2610.19605v1 Announce Type: cross Abstract: Long-output inference in large language models suffers from severe KV cache memory and bandwidth bottlenecks. While KV cache quantization alleviates t
arXiv:2610.19372v1 Announce Type: new Abstract: Runtime policy verifiers can prevent agents from executing disallowed actions, but their decisions are typically used only for enforcement and discarded
arXiv:2610.16145v1 Announce Type: new Abstract: Agent frameworks can persist checkpoints, trace tool calls, and limit resource use, but these facilities often assign different identities to execution,
arXiv:2610.19615v1 Announce Type: new Abstract: Reinforcement learning has enabled robust locomotion for legged robots, but learned policies are typically tied to a specific robot morphology and requi
arXiv:2409.00696v4 Announce Type: replace-cross Abstract: Rating-based human evaluation has become an essential tool to accurately evaluate the impressive performance of large language models (LLMs).
arXiv:2610.15969v1 Announce Type: new Abstract: In multi-agent reinforcement learning, exposing a cooperator to partners with diverse strategies during training is key to better collaboration at test
arXiv:2610.18917v1 Announce Type: cross Abstract: Urban venues such as restaurants, shops, bars and parks host the daily life of a city, and their collective rise and fall signals how vibrant a neighb
arXiv:2610.19143v1 Announce Type: new Abstract: While Vision-Language-Action models have enabled impressive cross-task transfer in zero-shot settings, adaptation methods are often necessary to achieve
arXiv:2610.20402v1 Announce Type: cross Abstract: Feldman's memorization lower bound assigns a frequency coefficient based on how often a point appears in the training sample. We give an exact three-p
arXiv:2610.14843v1 Announce Type: new Abstract: Claims of recursive self-improvement are easy to make and hard to falsify: did improvement come from re-using designer-provided mechanisms, or from mech
arXiv:2610.19048v1 Announce Type: new Abstract: Record linkage and demographic inference from names are routine demands in social science. Most approaches to these problems require labeled training da
arXiv:2610.18755v1 Announce Type: cross Abstract: Reinforcement learning (RL) from unit tests is the common recipe for post-training code models, but a unit test only asks whether the final program pa
arXiv:2610.14937v1 Announce Type: new Abstract: How do language models generalize from their training data? This is a central challenge for alignment: models reliably learn their training distribution
arXiv:2610.20288v1 Announce Type: new Abstract: Conditional generative models are increasingly used for scientific prediction beyond the direct training support. Yet scientific questions often concern
arXiv:2610.20587v1 Announce Type: cross Abstract: An agentic market is a decentralized serving-time market where AI agents bid to claim each user prompt. An agent's bid is based on its expected value
arXiv:2412.12877v3 Announce Type: replace Abstract: Recent Artificial Intelligence (AI)-based video editing has enabled users to edit videos through simple text prompts, significantly simplifying the
arXiv:2610.20538v1 Announce Type: new Abstract: Multimodal code intelligence connects the symbolic world of programs with the visual world their execution creates. Interest in this capability is burge
arXiv:2610.19762v1 Announce Type: new Abstract: Medical prescribing errors for multimorbid patients, who require complex drug regimens, are a major patient safety concern. These errors often stem from
arXiv:2610.16778v1 Announce Type: cross Abstract: Cloud graphical user interface (GUI) agents send instructions and screenshots to remote planners, exposing personally identifiable information (PII).
arXiv:2610.19336v1 Announce Type: new Abstract: Digital twins are increasingly used in robotics to simulate interactions, predict action outcomes, and support planning. However, constructing accurate
arXiv:2610.14912v1 Announce Type: new Abstract: Language models are increasingly used to create synthetic survey samples, also called silicon samples. Decision models are a new class of language model