🔍 搜索

Can Learning to Answer Spatial Questions Improve Novel-View Generation?

Generating novel views from limited observations requires preserving both the appearance and spatial structure of a scene. Although recent approaches exploit fine-grained geometric cues, they do not e...

Provably adaptive sampling with uniform and remasking discrete diffusion models

Discrete diffusion models offer a promising alternative to autoregressive generation by enabling parallel updates, but their sampling efficiency can depend strongly on the choice of the forward proces...

Estimating Flow Velocity and Vehicle Angle-of-Attack from Non-invasive Piezoelectric Structural Measurements Using Deep Learning

Accurate estimation of aerodynamic state variables such as freestream velocity and angle of attack (AoA) is important for aerodynamic load prediction, flight control, and model validation. This work p...

Enhancing Alignment for Unified Multimodal Models via Semantically-Grounded Supervision

Unified Multimodal Models (UMMs) have emerged as a promising paradigm that integrates multimodal understanding and generation within a unified modeling framework. However, current generative training ...

Retrieval Heads Meet Vision: Uncovering How VLMs Locate and Extract Visual Information

Vision-language models (VLMs) can locate an image region referred to by a text prompt and route the corresponding visual evidence to the output, yet the internal mechanism behind this behavior is not ...

CreativeInstruct: Scalably Teaching LLMs to Balance Quality, Creativity, and Diversity

While post-training improves the capabilities of large language models (LLMs), it generally lowers their output diversity and creativity, negatively impacting tasks that explicitly require creativity ...

Freeform Preference Learning for Robotic Manipulation

Reward design remains a central bottleneck for autonomous robot policy improvement, especially in long-horizon manipulation tasks where sparse success labels provide too little signal and binary prefe...

Algorithms for Robbins' Problem using Markov Decision Processes

In this paper, we consider Robbins' problem, which is a full information variant of the well-known secretary selection problem. In this version of the problem, the goal is to minimize the expected ran...

NASimJax: GPU-Accelerated Policy Learning Framework for Penetration Testing

Penetration testing, the practice of simulating cyberattacks to identify vulnerabilities, is a complex sequential decision-making task that is inherently partially observable and features large action...

GaussiAnimate: Reconstruct and Rig Animatable Categories with Level of Dynamics

Free-form bones, that conform closely to the surface, can effectively capture non-rigid deformations, but lack a kinematic structure necessary for intuitive control. Thus, we propose a Scaffold-Skin R...

Marine primary production enhances removal of hydrophobic organic contaminants from Eurasian Arctic shelf waters.

Global oceans store large quantities of human-made persistent hydrophobic organic contaminants (HOCs), driven by the organic carbon (OC) cycle. Despite diverse organic matter composition, bulk OC is t...

A fully marine stem amphibian from the Late Triassic of China.

Land vertebrates colonized the sea repeatedly through geologic time with various levels of marine adaptation, from partial use of marine resources to fully marine lifestyles. Mammals, reptiles and bir...

Turning the TIDE: Cross-Architecture Distillation for Diffusion Large Language Models

Diffusion large language models (dLLMs) offer parallel decoding and bidirectional context, but state-of-the-art dLLMs require billions of parameters for competitive performance. While existing distill...

Bacterial composition of the microbiome of mesopelagic fishes from the Northeast Atlantic.

The mesopelagic zone, between 100 and 1000 m depth, contains up to 90% of marine vertebrate biomass. Mesopelagic fishes are key components of these ecosystems through their feeding interactions and pl...

Host matters: coral reef fish species show distinct skin microbiome responses to upwelling-driven environmental changes.

Severe environmental disturbances compromise ecological thresholds in aquatic environments.

Large-scale disturbances can profoundly alter aquatic ecosystems by modifying water chemistry, sediment dynamics, and biodiversity across connected freshwater and marine environments. The 2015 Fundão ...

Sharp Phase Transition for the Formation of Infinite Tubes

Classical bond percolation theory studies the conditions for a given point in a random graph to be connected to infinity, or "escape" to infinity, via a sequence of random edges. In this work, we pres...

Query-Efficient Rejection Sampling for Parallel Masked Diffusion

Masked diffusion models (MDMs) accelerate generation by unmasking multiple tokens in parallel, but sampling these tokens independently ignores their correlations and biases the output. Anari et al. (2...

Minimising decompression and warming during deep seawater collection increases abundance and activity of autochthonous bacteria and archaea.

The deep ocean hosts autochthonous pressure-adapted microorganisms that are unique to this environment, as well as allochthonous pressure-sensitive members transported from shallow depths by vertical ...

Wound healing in Atlantic spiny dogfish sharks.

Field observations and limited experimental studies indicate that elasmobranchs can repair substantial skin injuries, but the temporal course and cellular composition of wound healing in Atlantic spin...