Search papers, labs, and topics across Lattice.
100 papers published across 2 labs.
Neural Cellular Automata, blending Wolfram's recursive programs with neural networks, offer a fresh perspective on modeling complex, self-organizing systems.
Thompson Sampling can be just as efficient with pairwise preference feedback as it is with scalar rewards, opening up new avenues for optimization in human-in-the-loop and experimental design scenarios.
Guaranteeing physical constraints in your ML model doesn't have to sacrifice uncertainty quantification – this Bayesian method bakes in linear equalities while shrinking credible intervals.
PINNs offer a promising new approach to solving complex problems in differential geometry by directly minimizing differential functionals.
Diffusion models, typically used for image generation, can now forecast infectious disease with accuracy rivaling traditional ensemble methods, offering a new tool for public health.
Neural Cellular Automata, blending Wolfram's recursive programs with neural networks, offer a fresh perspective on modeling complex, self-organizing systems.
Thompson Sampling can be just as efficient with pairwise preference feedback as it is with scalar rewards, opening up new avenues for optimization in human-in-the-loop and experimental design scenarios.
Guaranteeing physical constraints in your ML model doesn't have to sacrifice uncertainty quantification – this Bayesian method bakes in linear equalities while shrinking credible intervals.
PINNs offer a promising new approach to solving complex problems in differential geometry by directly minimizing differential functionals.
Diffusion models, typically used for image generation, can now forecast infectious disease with accuracy rivaling traditional ensemble methods, offering a new tool for public health.
Forget perturbation theory: HAML meta-learns effective qubit Hamiltonians directly from multi-mode simulations, enabling accurate characterization even when traditional methods break down.
Unlock the secrets hidden in your lab's backed-up microscopy data: style transfer networks can now "re-imagine" images as if they were captured with different instrument settings.
Frozen vision-language models can dramatically improve abnormality grounding in rare disease imaging by iteratively refining decisions through optimized instructions and visual perturbations.
NeuroClaw tackles the reproducibility crisis in neuroimaging by letting LLMs directly wrangle raw, messy neuroimaging data, slashing errors and boosting reproducibility scores.
Finally, a dataset exists to train and benchmark algorithms for automatically detecting airway bifurcations in 3D CT scans, a crucial step towards understanding respiratory diseases.
Scaling up pathology foundation models doesn't guarantee better survival prediction—a distilled model with 8% of the parameters can outperform its larger teacher.
Cytogeneticists can now slash chromosome analysis time from days to seconds with Aycromo, an open-source platform that democratizes access to high-performance deep learning models.
Kinks in aligned polymers don't just reduce thermal conductivity, they induce superdiffusive heat transport at long lengths, scaling conductivity with length as $L^{1/3}$.
Unlock accurate friction estimation for any material pairing with just a handful of proxy material measurements, slashing experimental costs.
AtomWorld achieves the previously impossible: simulating the degradation of reactor pressure vessel steel at the atomistic level across year-and-meter scales.
AI agents can autonomously orchestrate the entire machine learning pipeline for protein-protein interaction prediction, from data collection to rule induction, offering a new level of automation and interpretability.
Tensor networks, previously confined to lattice models, can now efficiently tackle continuous-space statistical mechanics problems, outperforming Monte Carlo in hard-disk simulations.
Quantum emitter arrays can exhibit nonlinear optical behavior in their linear spectra, opening new avenues for controlling light-matter interactions at the nanoscale.
Bridging the gap between simulation and reality, this work delivers machine learning potentials that are experimentally-validated, uncertainty-aware, and as accurate as the best electronic-structure methods.
Vib2Conf achieves unprecedented accuracy in identifying 3D molecular conformations from vibrational spectra, even distinguishing between near-isomeric conformers differing by only ~1 Å RMSD.
Unlock spectroscopic and electronic observables in large-scale molecular simulations by learning the electron density directly, paving the way for more comprehensive and transferable machine-learned interatomic potentials.
Doping a sodium-ion battery cathode with boron can significantly boost its capacity and stability, offering a promising route to improved energy storage.
Computation can now design light-activated drugs: a novel compound achieved a 15x boost in cancer target inhibition upon green light exposure.
Ditch the prompts: DiffuSAM adapts SAM2 for medical image segmentation by synthesizing mask embeddings with a diffusion model, achieving strong performance without fine-tuning or expert input.
Biophysically-constrained models of gene regulation, learned via probability flow matching, are the only ones that accurately predict cell fate decisions and responses to perturbations, even when other models interpolate the training data just as well.
Quantum kernels unlock signal in medical image embeddings where classical methods fail, suggesting a new path for extracting value from medical foundation models.
Species identification and discovery, traditionally treated as separate problems, can be unified into a single framework that leverages retrieval-augmented reasoning for improved accuracy and interpretability.
AI reveals that the true reaction pathway for retinal isomerization follows a surprising S-shaped trajectory driven by non-equilibrium dynamics, a feature invisible to traditional free-energy surface analysis.
Vertex corrections are essential for accurate $GW$ calculations of nuclear densities, revealing the limitations of standard self-energy approximations in electron-nuclei correlation.
Unlock the secrets of the deep: OceanPile, a massive, meticulously curated multimodal dataset, finally brings the power of foundation models to the vast and underexplored ocean.
Explicitly conditioning neural surrogates on supersaturation dramatically improves their accuracy in simulating crystal growth dynamics compared to implicit inference, especially with limited data.
Unlock reusable architectures for climate data super-resolution: a single diffusion model now handles spatial upscaling from 1x to 25x and temporal upscaling from 1x to 6x.
You can boost insurance claim prediction accuracy by combining simple environmental features with location data, even when you lack detailed individual-level spatial information.
Forget painstakingly tuning RL algorithms for quantum circuit optimization – smart replay buffer engineering alone can slash training time by up to 90% and boost sample efficiency by 32x.
Ditch KL divergence for IPMs in Bayesian experimental design and watch your credible sets tighten and your designs stabilize, even when your model's a bit off.
Uncover hidden GFlowNet training dynamics with GFlowState, a visual analytics tool that reveals how these models explore the sample space and shift sampling probabilities.
Unlock exponentially faster computation of dynamical system representations by exploiting the algebraic structure of Koopman eigenfunctions.
Quotient-space diffusion elegantly sidesteps the need to learn symmetry transformations, leading to more efficient and accurate generative models for systems with inherent symmetries.
Solve new PDEs 100x faster with 10x less error by learning a transferable PINN representation and adapting to new equations with a single closed-form calculation.
Persistent homology, when applied to eye-tracking data via novel filtration techniques, unlocks dyslexia detection performance exceeding traditional statistical methods.
Despite their architectural differences, Transformer-based genome language models can provide equally reliable biological insights as CNNs, as revealed by attention-based explainability methods.
ML models can accurately predict quantum properties out-of-distribution, but still fail to accelerate SCF convergence – until now.
LSTMs can bring low-cost air quality sensors up to regulatory compliance, unlocking dense urban monitoring networks previously limited by calibration challenges.
ResGIN-Att's cross-attention mechanism not only boosts drug synergy prediction but also offers a peek into the "why" behind drug interactions by highlighting crucial chemical substructures.
PINNs can now efficiently solve highly oscillatory wave equations in heterogeneous media, thanks to a Green's function-based integral formulation that cuts computation by 10x and avoids absorbing boundary layers.
Compact datasets in n-dimensional space can be transformed into linearly separable sets using diffeomorphisms and shallow, wide neural networks, challenging the need for complex architectures in certain classification tasks.
Quantum trajectory reversal, previously understood through specific feedback protocols, is now shown to be fundamentally linked to score-based diffusion, opening the door to ML-driven control in noisy, real-world quantum systems.
Simple neural networks can accurately emulate complex aerosol microphysics in climate models, but only with careful attention to scaling and training convergence.
Even with explicit long-range interactions, machine learning potentials still struggle to capture the subtle medium-range order in silica glass, hinting at fundamental limitations in how these models represent the liquid-to-glass transition.
Automating the semantic translation of research questions into scientific workflows slashes data transfer by 92% and keeps LLM overhead under 15 seconds per query.
A novel logic-based approach makes inferring complex, temporally-extended events from timestamped data tractable, even in the messy real-world of medical records.
Applying differential privacy to survival analysis can obliterate statistical significance and predictive power, even with relatively large datasets and optimistic clipping bounds.
H&E slides can now predict spatial gene expression with significantly improved accuracy and robustness, even when faced with unseen slide variations, thanks to a novel post-hoc calibration technique.
A single pneumatic input and clever use of magneto-elastic hysteresis can drive a surprisingly simple and effective peristaltic pump.
By adversarially removing camera-specific fingerprints, FryNet forces models to learn genuine chemical representations from thermal images, enabling robust and generalizable frying oil oxidation assessment.
By explicitly addressing often-overlooked fusion erasure errors, this new compilation scheme unlocks exponentially more robust photonic quantum computations.
Atom-scale simulations reveal how water content and ionomer distribution impact the charging behavior of Nafion thin films on platinum, offering design insights for improved electrocatalysis.
COFs can withstand defects surprisingly well: mechanical properties remain stable even with defects, but thermal conductivity plummets, revealing design trade-offs.
Forget standard models: a refined Judd-Ofelt theory unlocks more accurate predictions of Pr3+ luminescence, revealing pathways to boost laser performance across multiple wavelengths.
Surface-mediated autocatalysis isn't just about growth; this model reveals how surface interactions can tip the balance between explosive population growth and complete extinction.
Existing methods for quantifying molecular rotation break down when motion becomes complex, but this new method accurately captures rotational dynamics from fluid to solid states.
Skip the pixel-perfect annotations: attention-based MIL with pathology foundation models can predict lung cancer growth patterns from whole slide images with surprisingly high accuracy.
Lithology classification gets a reasoning upgrade: GeoMind's agentic workflow beats static methods by grounding decisions in geological evidence and constraints.
AI agents in medical research aren't ready for prime time: a new audit framework reveals that over half of evaluated skills fall below the "Limited Release" threshold, highlighting the need for domain-specific safeguards.
Finally, a meta-learning approach that uses readily available negative control samples can close the persistent domain gap in biomedical imaging, making deep learning models practically usable across different experimental batches.
A new global dataset reveals intricate deployment patterns and operational dynamics of offshore wind infrastructure, enabling unprecedented temporal analysis.
Gauge-equivariant GNNs unlock the ability to learn intrinsically nonlocal observables in lattice gauge theories by directly embedding non-Abelian symmetries into message passing.
Current evaluation metrics for trajectory inference can mislead researchers, but functional KL divergence offers a clearer, more reliable comparison of methods in sparse data conditions.
Synthesizing complete ice-layer data from sparse radar traces is now possible by conditioning on climate model outputs, enabling more accurate analysis of ice dynamics and snow accumulation.
Get 82x faster Bayesian inference for equipment monitoring by replacing MCMC with neural nets trained on simulated data.
Unlock 10x faster simulation-based inference in hierarchical models by training on single-site simulations and assembling synthetic multi-site data.
Whitening neuroimaging features can transform linear models from black boxes into interpretable tools for understanding brain pathology.
Sampling plausible configurations of digital twins can reveal multiple valid parameterizations, enhancing model adaptation in complex natural systems.
Imagine slashing the human effort needed to go from hypothesis to submission-ready ML theory paper by orders of magnitude.
GNNs can slash storm surge forecast errors by over 70%, offering a faster and more accurate alternative to traditional numerical models for coastal disaster prediction.
Identifying causal effects can now be achieved in quasi-polynomial time, transforming the feasibility of causal inference in complex datasets.
Unlock the secrets of AI weather models: a new tool reveals how latent representations encode interpretable meteorological features.
AI models can generate visually convincing yet physically implausible fluid dynamics solutions, revealing a critical flaw in their design.
Ditch the expensive energy calculations: this new ML-DFT approach learns directly from ground-state densities, achieving state-of-the-art accuracy with improved runtime scaling.
CDLF outperforms traditional forecasting methods by adapting to new product data in real-time, even in the absence of historical outcomes.
SPD-SheafNets learn richer geometric representations than standard GNNs by operating directly on matrix-valued features, achieving SOTA on molecular property prediction by capturing relationships between directions.
Forget one-shot generation: Mol-Debate's iterative debate loop unlocks state-of-the-art molecular design by dynamically reconciling semantic intent with structural feasibility.
Achieve more reliable and interpretable virtual cell perturbation predictions by combining knowledge-driven multimodal modeling with evidence retrieval.
Finally, a deep learning model for AKI prediction that doesn't just predict, but tells you *why*, by tracing the causal chain of physiological events.
Energy-dissipation principles can revolutionize how we infer potential functions in noisy, incomplete data environments, achieving remarkable robustness in generalized diffusion processes.
Force-feeding physics to LSTMs slashes battery thermal runaway prediction errors by over 80%, making your next e-bike less likely to explode.
Guaranteeing symmetrizable hyperbolicity in machine-learned radiative transfer models is now possible in 2D2V, opening the door to stable and accurate simulations in more complex geometries.
Forget finite differences: Fourier Weak SINDy offers a derivative-free approach to system identification that's robust to noise and leverages spectral estimation for interpretable equation learning.
Automated identification of individual animals can only be effective if it aligns with ecological questions and data practices, not just algorithmic accuracy.
Research ideas generated through a novel multi-agent system show a significant boost in diversity and novelty, outperforming traditional LLM methods.
LLMs can autonomously navigate the notoriously complex task of alloy phase diagram construction, outperforming traditional ML methods and even exhibiting complementary strengths when combined with domain-specific models.
MambaLiteUNet achieves state-of-the-art skin lesion segmentation with 93% fewer parameters and 97% fewer GFLOPs than U-Net, proving that efficient architectures can outperform traditional models in medical imaging.
LLMs that play nice in behavioral economics games make better AI scientists, suggesting cooperation isn't just about general smarts.
Forget prompting LLMs to directly predict hundreds of fields: a two-stage approach with a stable intermediate JSON summary and a deterministic compiler achieves strong performance on CRF filling while being language-agnostic.
Discover expertise and collaborators in battery research at a global scale, grounded in semantic understanding rather than just citations.
Stabilizing keyframe localization in fetal ultrasound can be achieved by temporally aggregating structure-augmented planes, leading to more reliable biometric measurements.
Even with severely degraded input, RetinaDiff unlocks reliable retinal blood flow imaging from just a handful of frames, outperforming conventional methods.
Repurposing routine knee X-rays for AI-powered bone-loss screening could dramatically expand access to early osteoporosis detection.
Quantum computers can serve as effective "topographical preconditioners" to guide classical solvers in high-dimensional optimization, bypassing the limitations of both purely quantum and classical approaches.
Convex DFT fixes a long-standing flaw in DFT, enabling accurate and efficient simulations of photochemistry by restoring the correct topology near conical intersections.
