Ensembits is the first tokenizer of protein conformational ensembles that outperforms static tokenizers on RMSF prediction and matches them on function and mutation tasks while using less pretraining data.
super hub Mixed citations
Jumper , author R
Mixed citation behavior. Most common role is background (64%).
53
Pith papers citing it
40.9k
external citations · Crossref
Background
64% of classified citations
hub tools
citation-role summary
background 8
method 2
other 1
citation-polarity summary
authors
co-cited works
representative citing papers
HO-FNO extends standard FNO with n-linear spectral mixing and shows improved accuracy on nonlinear PDE benchmarks, sometimes with a single layer beating deeper FNO models.
PhySciBench benchmark shows current AI models achieve at most 33.5% accuracy on physical science tasks; DelveAgent framework improves accuracy by up to 7.5 points and cuts costs to one-third.
AMPGAN v3 generates non-canonical AMPs with D-amino acids and modifications using two discriminators for stability, validated with two active candidates in vitro, alongside the PepCraft multi-agent discovery framework.
RATrain introduces a resource-aware scheduler and MT-3000-specific backend for 1F1B LLM training that achieves 1.35x speedup and 97% scaling efficiency while preserving training correctness.
DPA4 is a new SE(3)-equivariant interatomic potential with EMFA SO(2) convolution that sets new accuracy-cost records on Matbench Discovery and SPICE benchmarks using fewer parameters than prior models.
Presents a general framework for generator matching on projected image spaces from latent Markov processes, generalizing static latent results to dynamic conditional processes.
Derives a conditional-marginal entropy-rate objective for bridge-aware discretization that yields U-shaped schedules and improves low-NFE sample quality on 2D, CIFAR-10, and protein tasks.
The paper introduces compositional interpretability as a category-theoretic framework that casts mechanistic explanations as commuting syntactic-semantic mappings optimized under faithfulness and complexity constraints derived from minimum description length.
Masked-position MLM plus JEPA latent prediction outperforms MLM-only pretraining on 10-11 of 16 downstream tasks for 35M-150M protein models while JEPA alone fails.
TCD-Arena is a new customizable testing framework that runs millions of experiments to map how 33 different assumption violations affect time series causal discovery methods and shows ensembles can boost overall robustness.
SMC forgets its initial condition geometrically in the jump chain and as 1/ℓ in continuous genetic distance, justifying independent-locus approximations.
Dual Triangle Attention achieves effective bidirectional attention with built-in positional inductive bias via dual triangular masks, outperforming standard bidirectional attention on position-sensitive tasks and showing strong masked language modeling results with or without positional embeddings.
DenseAMs show tradeoffs between entropy production, retrieval accuracy, and speed at intermediate loads, with a new failure mode in higher-order networks at finite temperature.
Quantum circuits for coherent multilayer neural network inference achieve quadratic to polylogarithmic speedups over classical methods depending on quantum data access models for inputs and weights.
AlphaEvolve is an LLM-orchestrated evolutionary coding agent that discovered a 4x4 complex matrix multiplication algorithm using 48 scalar multiplications, the first improvement over Strassen's algorithm in 56 years, plus optimizations for Google data centers and hardware.
A novel neural architecture based on Pairformer is introduced for learning committor functions to better capture dynamical features in biomolecular rare events without specialized priors.
The work constructs a permutation-equivariant quantum GNN that implements message passing at selectable Weisfeiler-Leman levels, supports pre-training on small graphs, and demonstrates readout scalability with simulations up to 56 qubits on synthetic, molecular, and TSP datasets.
Local 2- and 3-cycles enhance RNN computational capacity for Boolean functions, predicted by structural statistics, while adding interneurons boosts large networks.
JEDEL maps pharmacophore patterns to scalable combinatorial synthesis routes for DNA-encoded libraries, producing focused libraries that outperform baselines on 18 targets in zero-shot mode.
The α-index is a conserved position-weighted authorship framework with a senior-author penalty that decreases credit as the number of middle authors increases.
Early-exit GNNs for link prediction move the speed-quality Pareto frontier on the HeaRT benchmark by allowing implicit early exiting without auxiliary losses.
New class of sequence kernels for Gaussian processes that use substitution matrices and local linearity to enable data-efficient prediction of protein properties, with extensions to structure-aware multi-task learning.
PDE-Agents shows a LangGraph-orchestrated multi-agent LLM framework with GraphRAG that reaches 100% task success and perfect material fidelity on novel materials in ablation tests, with 97.8% success across 1369 production runs.
citing papers explorer
-
ENSEMBITS: an alphabet of protein conformational ensembles
Ensembits is the first tokenizer of protein conformational ensembles that outperforms static tokenizers on RMSF prediction and matches them on function and mutation tasks while using less pretraining data.
-
Higher-Order Fourier Neural Operator: Explicit Mode Mixer for Nonlinear PDEs
HO-FNO extends standard FNO with n-linear spectral mixing and shows improved accuracy on nonlinear PDE benchmarks, sometimes with a single layer beating deeper FNO models.
-
Deep Research in Physical Sciences: A Multi-Agent Framework and Comprehensive Benchmark
PhySciBench benchmark shows current AI models achieve at most 33.5% accuracy on physical science tasks; DelveAgent framework improves accuracy by up to 7.5 points and cuts costs to one-third.
-
Agentic Discovery of Non-Canonical Antimicrobial Peptides with AMPGAN v3
AMPGAN v3 generates non-canonical AMPs with D-amino acids and modifications using two discriminators for stability, validated with two active candidates in vitro, alongside the PepCraft multi-agent discovery framework.
-
RATrain: A Resource-Aware Training Runtime for Large Language Models on Bandwidth-Constrained Heterogeneous Supercomputing Platforms
RATrain introduces a resource-aware scheduler and MT-3000-specific backend for 1F1B LLM training that achieves 1.35x speedup and 97% scaling efficiency while preserving training correctness.
-
DPA4: Pushing the Accuracy-Cost Frontier of Interatomic Potentials with EMFA SO(2) Convolution
DPA4 is a new SE(3)-equivariant interatomic potential with EMFA SO(2) convolution that sets new accuracy-cost records on Matbench Discovery and SPICE benchmarks using fewer parameters than prior models.
-
Latent Process Generator Matching
Presents a general framework for generator matching on projected image spaces from latent Markov processes, generalizing static latent results to dynamic conditional processes.
-
Entropy Across the Bridge: Conditional-Marginal Discretization for Flow and Schr\"odinger Samplers
Derives a conditional-marginal entropy-rate objective for bridge-aware discretization that yields U-shaped schedules and improves low-NFE sample quality on 2D, CIFAR-10, and protein tasks.
-
From Mechanistic to Compositional Interpretability
The paper introduces compositional interpretability as a category-theoretic framework that casts mechanistic explanations as commuting syntactic-semantic mappings optimized under faithfulness and complexity constraints derived from minimum description length.
-
ProteinJEPA: Latent prediction complements protein language models
Masked-position MLM plus JEPA latent prediction outperforms MLM-only pretraining on 10-11 of 16 downstream tasks for 35M-150M protein models while JEPA alone fails.
-
TCD-Arena: Assessing Robustness of Time Series Causal Discovery Methods Against Assumption Violations
TCD-Arena is a new customizable testing framework that runs millions of experiments to map how 33 different assumption violations affect time series causal discovery methods and shows ensembles can boost overall robustness.
-
Rates of forgetting for the sequentially Markov coalescent
SMC forgets its initial condition geometrically in the jump chain and as 1/ℓ in continuous genetic distance, justifying independent-locus approximations.
-
Dual Triangle Attention: Effective Bidirectional Attention Without Positional Embeddings
Dual Triangle Attention achieves effective bidirectional attention with built-in positional inductive bias via dual triangular masks, outperforming standard bidirectional attention on position-sensitive tasks and showing strong masked language modeling results with or without positional embeddings.
-
Stochastic Thermodynamics of Associative Memory
DenseAMs show tradeoffs between entropy production, retrieval accuracy, and speed at intermediate loads, with a new failure mode in higher-order networks at finite temperature.
-
Accelerating Inference for Multilayer Neural Networks with Quantum Computers
Quantum circuits for coherent multilayer neural network inference achieve quadratic to polylogarithmic speedups over classical methods depending on quantum data access models for inputs and weights.
-
AlphaEvolve: A coding agent for scientific and algorithmic discovery
AlphaEvolve is an LLM-orchestrated evolutionary coding agent that discovered a 4x4 complex matrix multiplication algorithm using 48 scalar multiplications, the first improvement over Strassen's algorithm in 56 years, plus optimizations for Google data centers and hardware.
-
Navigating committor landscape of biomolecules with a general pairwise interaction model
A novel neural architecture based on Pairformer is introduced for learning committor functions to better capture dynamical features in biomolecular rare events without specialized priors.
-
Scalable Message-Passing Quantum Graph Neural Networks in the Weisfeiler-Leman Hierarchy
The work constructs a permutation-equivariant quantum GNN that implements message passing at selectable Weisfeiler-Leman levels, supports pre-training on small graphs, and demonstrates readout scalability with simulations up to 56 qubits on synthetic, molecular, and TSP datasets.
-
Identifying structural design principles shaping the computational abilities of recurrent neural networks
Local 2- and 3-cycles enhance RNN computational capacity for Boolean functions, predicted by structural statistics, while adding interneurons boosts large networks.
-
JEDEL: Zero-Shot DNA-Encoded Library Design for Early-Stage Drug Discovery
JEDEL maps pharmacophore patterns to scalable combinatorial synthesis routes for DNA-encoded libraries, producing focused libraries that outperform baselines on 18 targets in zero-shot mode.
-
The $\alpha$-Index: A Penalized Authorship-Integrity Framework for Position-Weighted Scientific Contribution
The α-index is a conserved position-weighted authorship framework with a senior-author penalty that decreases credit as the number of middle authors increases.
-
Early-Exit Graph Neural Networks for Link Prediction
Early-exit GNNs for link prediction move the speed-quality Pareto frontier on the HeaRT benchmark by allowing implicit early exiting without auxiliary losses.
-
Flexible Kernels for Protein Property Prediction
New class of sequence kernels for Gaussian processes that use substitution matrices and local linearity to enable data-efficient prediction of protein properties, with extensions to structure-aware multi-task learning.
-
PDE-Agents: An LLM-Orchestrated Multi-Agent Framework for Automated Finite Element Simulations with Knowledge Graph-Augmented Reasoning
PDE-Agents shows a LangGraph-orchestrated multi-agent LLM framework with GraphRAG that reaches 100% task success and perfect material fidelity on novel materials in ablation tests, with 97.8% success across 1369 production runs.
-
Methods for Inferring Interaction Potentials from Cross-Linking Mass Spectrometry Data
Develops and tests algorithms adapting inverse Henderson problem solvers to parameterize multi-component interaction potentials from XL-MS data in homogeneous and three-phase systems.
-
Towards Understanding Self-Pretraining for Sequence Classification
Self-pretraining improves Transformer sequence classification by enabling learning of proximity-biased attention from positional encodings that label supervision alone cannot easily acquire from random starts.
-
CrystalBoltz: End-to-End Protein Structure Determination via Experiment-Guided Diffusion for X-Ray Crystallography
CrystalBoltz performs experiment-guided posterior sampling with diffusion models on structure-factor amplitudes for protein structure determination, reporting lower RMSD and R-factors than baselines with 33x faster runtime.
-
NOVA: Fundamental Limits of Knowledge Discovery Through AI
NOVA models the generate-verify-accumulate-retrain loop and proves cumulative discovery cost scales as Theta(c_gen D^alpha) under Zipf tail equivalence with alpha greater than 1.
-
ShardTensor: Domain Parallelism for Scientific Machine Learning
ShardTensor is a domain-parallelism system for SciML that enables flexible scaling of extreme-resolution spatial datasets by removing the constraint of batch size one per device.
-
Supercharging Bayesian Inference with Reliable AI-Informed Priors
Rectified AI priors, obtained by correcting AI-induced data laws before embedding them in techniques like Dirichlet process priors, reduce bias, improve credible interval coverage, and boost performance in tasks like skin disease classification.
-
A physics-informed neural network approach to solve the spatially inhomogeneous electron Boltzmann equation
A specialized PINN architecture solves the spatially inhomogeneous electron Boltzmann equation with high accuracy across gases and electric field strengths without case-specific tuning.
-
Flashlight: PyTorch Compiler Extensions to Accelerate Attention Variants
Flashlight is a compiler-native PyTorch framework that generates efficient fused kernels for arbitrary and data-dependent attention variants, supporting more cases than FlexAttention with competitive performance.
-
Fast and Interpretable Protein Substructure Alignment via Optimal Transport
PLASMA applies regularized optimal transport with Sinkhorn iterations to produce fast, interpretable residue-level alignments and similarity scores between protein structures.
-
Improving Inverse Folding for Peptide Design with Diversity-regularized Direct Preference Optimization
Diversity-regularized DPO fine-tuning of ProteinMPNN improves structural similarity scores by at least 8% over base model and sequence diversity by up to 20% over standard DPO for peptide inverse folding on OpenFold structures.
-
HSAP: A Hierarchical Sequence-aware Parallelism for Hybrid-Context Generative Models
HSAP introduces a hierarchical framework and sequence-aware algorithm with JIT-optimized NCCL communication to enable correct causal attention computation on hybrid-context packed sequences without limiting parallelism.
-
The FIL Hypothesis: Inductive Biases Help with Kernel Engineering
The FIL Hypothesis claims that inductive biases outperform purely data-driven methods on GPU programming tasks with non-trivial feedback loops.
-
Multiscale reconstruction of protein conformations from cryo-EM images
A multiscale optimization method using explicit protein backbone geometry reconstructs atomic models from cryo-EM data, showing improved RMSD and TM scores on three simulated datasets.
-
Instrumented data for causal scientific machine learning
Instrumented data augments observations with mechanistic models, uncertainty, and counterfactuals to enable causal interventions via Pearl's do-operator in scientific machine learning.
-
DeltaDiff: Training-Free, Physics-Guided Machine Learning for Predicting Mutant Protein Structures
DeltaDiff is a physics-guided inference method that predicts mutant protein structures from a baseline diffusion model without retraining, tested on three systems with nonlocal changes.
-
MOSAIC: Efficient Mixture-of-Agent Scheduling via Adaptive Aggregation and Inference Concurrency
MOSAIC uses an Integer Linear Program scheduler for expert placement and prompt assignment plus adaptive aggregation to achieve 1.7-2.3x end-to-end speedup on 4-GPU MoA workloads while keeping accuracy within 0.1pp.
-
AIBuildAI-2: A Knowledge-Enhanced Agent for Automatically Building AI Models
AIBuildAI-2 introduces a knowledge-enhanced agent with a hierarchical evolving external knowledge base that dynamically loads relevant AI development expertise, achieving first place on MLE-Bench at 70.7% medal rate.
-
Enabling Structure-Only Initialization and Out-of-Distribution Generalization in GNN-based Molecular Dynamics Simulators
GNN-based MD simulators achieve stable structure-only initialization and reliable OOD generalization through inference-time physics optimization and a GNN barostat on elastic network compression tasks.
-
Experiment-as-Code Labs: A Declarative Stack for AI-Driven Scientific Discovery
The paper introduces Experiment-as-Code Labs as a declarative stack synthesizing AI agents, systems orchestration, and physical lab control for AI-driven discovery.
-
Benchmarking open-source tools for in silico antiviral drug discovery
Boltz-2 and fine-tuned DrugFormDTA lead ML-based binding prediction while GNINA leads docking tools on a cleaned antiviral dataset, with performance varying by viral protein.
-
MIRA: A Score for Conditional Distribution Accuracy and Model Comparison
MIRA is a new analytic score for conditional distribution accuracy derived from equal probability mass assignment, enabling Bayesian model comparison via direct posterior validation.
-
Sampling Parallelism for Fast and Efficient Bayesian Learning
Sampling parallelism distributes Bayesian sample evaluations across GPUs for near-perfect scaling, lower memory use, and faster convergence via per-GPU data augmentations, outperforming pure data parallelism in diversity.
-
Galactica: A Large Language Model for Science
Galactica, a science-specialized LLM, reports higher scores than GPT-3, Chinchilla, and PaLM on LaTeX knowledge, mathematical reasoning, and medical QA benchmarks while outperforming general models on BIG-bench.
-
Building Digital Societies as Ecosystems: How Recognition and Repeat Relationships Sustain Cross-Community Work in Open Source
Cross-boundary collaboration in open source is sustained by a thin carrier layer of contributors and repeat relationships that increase pull request acceptance rates from 42% to 87%.
-
AIMBio-Mat: An AI-Native FAIR Platform for Closed-Loop Materials Discovery and Biomedical Translation
AIMBio-Mat is a conceptual blueprint for an AI-native, FAIR, governance-aware decision layer that formulates biomedical-materials discovery as constrained multi-objective optimization under uncertainty.
-
The New Associationism: Lessons from Deep Learning
Supervised learning across AI systems vindicates a uniform error-driven associationism for cognition, though operating inside advanced computational structures beyond classical associationist models.