Knowledge catalogue
arXiv:2609.32109v1 Announce Type: new Abstract: We study whether benchmarks reliably guide model selection as developers adapt to evaluation feedback across multiple criteria. We find that the worst-c
arXiv:2609.32346v1 Announce Type: new Abstract: This paper compares three human motion prediction methods - GMM-based clustering, TransFusion, and Graph-Mixer - evaluating their effectiveness in human
arXiv:2609.34262v1 Announce Type: cross Abstract: Agentic benchmarks guide model selection and training. Yet an agent can pass a task without demonstrating the intended capability. Such outcomes const
arXiv:2609.32656v1 Announce Type: cross Abstract: Multivariate Time Series Forecasting (MTSF) models that mix information across channels assume that the past of one channel carries information about
arXiv:2512.06983v2 Announce Type: replace-cross Abstract: World models enable agents to plan within imagined environments by predicting future states conditioned on past observations and actions. Howe
arXiv:2609.34301v1 Announce Type: new Abstract: Drug design couples property evaluation, conditional generation, structure-based design, and local optimization, yet machine learning systems typically
OpenAI: OpenAI is adopting a structured “safety case” documentation framework modeled after industries like aviation and nuclear power to govern frontier RL training — Loading... We believe we are ent
arXiv:2609.32810v1 Announce Type: cross Abstract: Multidisciplinary tumor boards integrate multimodal clinical observations and longitudinal patient histories through specialist discussions, yet bench
arXiv:2609.32652v1 Announce Type: cross Abstract: We study when geometry-induced soft state abstractions admit accurate finite-dimensional linear dynamics. Each state is represented by simplex-valued
arXiv:2609.32188v1 Announce Type: new Abstract: Text-to-image (TTI) models increasingly generate high-quality images from natural-language prompts, yet figurative language exposes a failure: a vehicle
arXiv:2609.33461v1 Announce Type: cross Abstract: Runtime monitoring traditionally evaluates a specification that is fixed before execution or externally modified when requirements change. In learning
arXiv:2609.33638v1 Announce Type: new Abstract: We introduce RareTrap, a framework for estimating the probability of severe behaviors in black box large language models (LLMs). A key challenge for pro
arXiv:2609.35349v1 Announce Type: new Abstract: The evaluation cost of transformers with softmax attention scales quadratically with sequence length. Kernel attention addresses this by replacing softm
We evaluate several models on 100 tasks from the [internal Binary Exploitation benchmark] (selected at random), and find that GLM-5.3 develops full control flow hijacks in 4% of the trials; Claude Myt
arXiv:2609.34179v1 Announce Type: cross Abstract: Retrieval-augmented generation systems are extensively instrumented with metrics, benchmarks, traces, and automated judges, but these tools do not dec
arXiv:2609.34163v1 Announce Type: new Abstract: Vision-language navigation (VLN) is typically evaluated as a one-way task, although deployed robots may need to return after reaching a goal. We study c
arXiv:2609.31694v1 Announce Type: new Abstract: Removing an invisible watermark and concealing the forensic evidence are distinct objectives: successfully disrupting the embedded watermark does not im
arXiv:2609.35717v1 Announce Type: new Abstract: Robots with kinematic loops, coupled actuators, and changing contacts require consistent models of configuration, motion, force, and dynamics. Yet these
arXiv:2609.34384v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models have shown strong capabilities in robotic manipulation, yet existing benchmarks typically assume that task-relevan
arXiv:2609.32036v1 Announce Type: new Abstract: We introduce ScreenHaystack, a dynamic needle-in-a-haystack benchmark for evaluating spatial reliability in GUI grounding. Instead of testing each targe
arXiv:2511.15565v3 Announce Type: replace Abstract: Human pose forecasting predicts future human poses from past observations for applications including action recognition, sports analysis, and human-
arXiv:2609.34736v1 Announce Type: cross Abstract: Large language model (LLM) routing aims to select the most suitable model for each incoming query. Most existing routers learn this decision directly
arXiv:2609.34153v1 Announce Type: new Abstract: Proteins change shape as they function, yet most inverse folding models predict amino acid sequences from a single, fixed backbone. A central challenge
arXiv:2609.33627v1 Announce Type: new Abstract: Despite recent progress in agentic multi-shot video generation, producing coherent and consistent long-form stories remains challenging. Existing agenti
arXiv:2609.32519v1 Announce Type: cross Abstract: Model steering can strengthen a target behavior while degrading other useful behaviors. We introduce Supervised Transcoder Replacement (STR) to reduce
arXiv:2606.15632v3 Announce Type: replace Abstract: Long videos challenge open-world segmentation systems to continually discover objects and preserve their identities through disappearance and reappe
arXiv:2609.33419v1 Announce Type: new Abstract: Comparisons in video self-supervised learning often evaluate complete training recipes rather than isolating the method itself: architecture, objective,
We let open models access the outside world during evals on tasks from public benchmarks. Here's what happened: Most went and found upstream commits containing fixes for the tasks they were supposed t
arXiv:2609.32532v1 Announce Type: new Abstract: A generalization gap in reinforcement learning (return on training levels minus return on held-out levels) is usually reported without a reference point
arXiv:2609.32677v1 Announce Type: cross Abstract: Self-improving agents increasingly rely on proxy verifiers to choose policy updates, yet deployment can change the world in which those updates are ev
arXiv:2603.13688v2 Announce Type: replace-cross Abstract: AI systems increasingly assist decision making by producing cheap factor-level assessments of complex inputs, but these assessments can be bia
arXiv:2609.34849v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards provides a sparse post-training signal: a single binary outcome evaluates the entire rollout, and every t
arXiv:2609.34440v1 Announce Type: new Abstract: Driving benchmark scores are increasingly used not only for evaluation but also as optimization targets. This raises a fundamental question: do score ga
arXiv:2609.30466v1 Announce Type: cross Abstract: Everyday Extended Reality (XR) systems aim to provide context-aware access to the right functionalities at the right time and place, with minimal manu
arXiv:2609.30292v1 Announce Type: cross Abstract: Online reviews shape consumer decisions, platform governance, and corporate reputation.Fake reviews compromise this information channel by injecting d
arXiv:2604.05224v2 Announce Type: replace Abstract: As Large Language Models (LLMs) are increasingly used to support search and information retrieval, it is critical that they accurately attribute con
arXiv:2512.06902v2 Announce Type: replace-cross Abstract: As software systems evolve, developers increasingly work across multiple programming languages and often face the need to migrate code from on
arXiv:2609.31074v1 Announce Type: new Abstract: Resource constraints make high-dimensional hyperspectral imaging challenging in autonomous perception, motivating the use of band selection methods. How
arXiv:2605.05725v2 Announce Type: replace Abstract: Time-series anomaly detection often returns scores or intervals, while analysts need to understand the abnormal behavior and the evidence supporting
arXiv:2609.31450v1 Announce Type: cross Abstract: Medical vision-language model (VLM) post-training is commonly evaluated through answer accuracy. We examine how changes in accuracy and training objec
arXiv:2609.31247v1 Announce Type: cross Abstract: Novel view synthesis (NVS) models can produce realistic new views of the same scene from different viewpoints. However, these generated views are not
arXiv:2609.30326v1 Announce Type: new Abstract: Activation steering changes a model's internal activations during inference without updating its weights, but a useful intervention must determine both
arXiv:2609.30763v1 Announce Type: new Abstract: Biomedical language models (LMs) encode textual semantics but do not explicitly preserve medical code hierarchies. We present Hyperbolic Clinical Ontolo
arXiv:2609.30288v1 Announce Type: new Abstract: In Transformer-based masked language models, attention is the primary mechanism for context mixing, but there are other ways to mix data across tokens.
arXiv:2504.08596v3 Announce Type: replace-cross Abstract: Hallucination, the generation of non factual content by AI systems, poses serious risks in medical contexts, where errors can directly affect
arXiv:2609.30781v1 Announce Type: new Abstract: Clinical measurements are recorded for some patients but not others, at rates that differ across hospitals, and marginal conformal coverage does not ens
arXiv:2609.30688v1 Announce Type: cross Abstract: This study conducts a systematic hyperparameter search across five deep learning architectures, DLinear, LSTM, PatchTST, ResAttLstm, and Mamba2, and n
OpenAI has paused all internal training and inference for its most capable frontier models after a misalignment incident in which an agent exploited a DNS filtering gap during routine research, attemp
arXiv:2608.25097v2 Announce Type: replace Abstract: Understanding how (multimodal) large language models perform on physics problems requires benchmarks that reflect the difficulty and breadth of expe
arXiv:2609.30657v1 Announce Type: cross Abstract: Large language model email assistants are particularly vulnerable to indirect prompt injection because untrusted email content can be retrieved into t
arXiv:2609.31505v1 Announce Type: new Abstract: Despite the growing capabilities of large language models (LLMs), prompt design remains largely heuristic and ad hoc. This project will explore extit{pr
arXiv:2606.06393v2 Announce Type: replace Abstract: Proper scoring rules provide a rigorous theoretical basis for the training and evaluation of probabilistic forecasts. In survival analysis, such for
arXiv:2609.30836v1 Announce Type: new Abstract: Deploying small language models (SLMs) on offline, resource-constrained edge devices such as remote sensing satellites presents a fundamental challenge:
arXiv:2609.31245v1 Announce Type: new Abstract: Individuals turn to large language models (LLMs) for guidance across a wide range of economic tasks, from comparing loan options and planning savings to
arXiv:2609.31164v1 Announce Type: cross Abstract: Recommender systems engineer serendipity to foster active exploration and break predictable consumption cycles. The problem with existing offline beyo
arXiv:2609.30969v1 Announce Type: new Abstract: Tactile information is essential for contact-rich manipulation tasks in robotics. Vision-based tactile sensors make it particularly easy to design end-t
arXiv:2609.31451v1 Announce Type: cross Abstract: The growing popularity of short videos has driven demand for one-click content creation. Visual templates turn uploaded images into personalized conte
arXiv:2609.30878v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) provides dense teacher targets, but evaluates them only along student-sampled rollouts. When the privileged teacher f
arXiv:2609.31463v1 Announce Type: cross Abstract: Infant movement analysis provides valuable biomarkers for the early identification of neurodevelopmental disorders. Recent advances in deep learning h
arXiv:2609.31076v1 Announce Type: new Abstract: Language agents struggle to act and learn in environments that require long sequences of low-level actions. Code-based abstractions can make these agent