diffusion models

Loneliness to consumption intention: emotional bonding and escapism in AI emotional companion app use

With the diffusion of generative AI, AI emotional companion apps have emerged as a new domain of digital consumption in which users may pay to extend relationship-like experiences. This study examines how loneliness and actual-ideal self-discrepancy are associated with experience-enhancing consumption intention and con...

Lookahead-Then-Verify: Reliable Constrained Decoding for Diffusion LLMs under Context-Free Grammars

Diffusion Large Language Models (dLLMs) have demonstrated promising capabilities and are increasingly used to produce formal languages defined by context-free grammars, such as source code and chemical expressions. However, as probabilistic models, they still struggle to generate syntactically valid outputs reliably. A...

A Grid of Model Contribution Functions and Spectra: to help with the interpretation of time-resolved observations of brown dwarfs

OVERVIEW This Zenodo record provides a grid of thermal emission spectra and their corresponding thermal-emission contribution-function summaries (average emission pressures indicating where the observed flux originates) for non-irradiated substellar objects in chemical equilibrium, as described in Phillippe et al. (202...

A Grid of Model Contribution Functions and Spectra: to help with the interpretation of time-resolved observations of brown dwarfs

OVERVIEW This Zenodo record provides a grid of thermal emission spectra and their corresponding thermal-emission contribution-function summaries (average emission pressures indicating where the observed flux originates) for non-irradiated substellar objects in chemical equilibrium, as described in Phillippe et al. (202...

BridgeDiff: A two-stage conditional diffusion model for bridge point clouds completion

Adaptive Mean Flow for Responsive Closed-Loop Robot Control

Diffusion- and flow-based robot policies have recently become widespread in robotic Imitation Learning (IL) due to their high performance and ability to model continuous and multimodal distributions. However, the iterative denoising procedure used by these models introduces significant prediction latency, hindering hig...

Physically Aware Generative Design for Spatial Mechanism Synthesis: A Differentiable Diffusion-Refinement Framework

Spatial mechanism synthesis requires joint reasoning over discrete topology and continuous geometry under strict loop-closure constraints. Existing search-based methods are often tied to predefined topology spaces, while purely data-driven generators can produce topologically plausible but kinematically infeasible mech...

Research on Green Technology Investment Strategies in Biodegradable Materials Supply Chain under Information Asymmetry

Scientific design of patent licensing contract and sales channel selection are key to breaking the green technology promotion dilemma of biodegradable materials. Under market scale information asymmetry, we construct a three-echelon supply chain principal agent model including a patent technology supplier, a biodegrada...

What deep learning can and cannot (yet) do for enzyme design.

Deep learning has substantially advanced structure-based enzyme design, yet reliably creating highly efficient enzymes remains out of reach. One emerging enzyme design strategy combines diffusion-based scaffold generation around active-site models with sequence design. Such structure-based approaches have expanded the...

Multi-Scale Mass Transfer Mechanism and Device Parameter Optimization in TCM Extraction: A Non-Equilibrium Thermodynamic and Fractional Diffusion Approach

Dataset for multi‑scale mass‑transfer modelling of traditional Chinese medicine extraction based on non‑equilibrium thermodynamics and fractional diffusion. Contains experimental extraction kinetic data for *Daphniphyllum calycinum, Astragalus membranaceus, Salvia miltiorrhiza, Pueraria lobata*, porosity‑permeability v...

Airflow velocity attenuation characteristics and a zoned dust control method in ventilation for drill-and-blast construction of V-shaped cross-sea tunnels

To address the problem of low dust removal efficiency in long-distance forced ventilation during drill-and-blast excavation of V-shaped subsea tunnels, which is caused by air leakage, along-path frictional resistance, and longitudinal slopes, this study takes the Qiongzhou Strait subsea tunnel project as the engineerin...

(Quantum)Annihilation Field Universe Model: From Particle Emergence to Cumulative Redshift

(Quantum) Annihilation Field Universe Model: From Particle Emergence to Cumulative Redshift Abstract This paper proposes an annihilation field universe model. The annihilation field is the core concept of this paper—a fundamental field in vacuum that forms localized excitations under the boundary conditions of quantum...