NLI autonomously discovers a vocabulary of primitive operations and interprets variable-length programs via a neural executor, allowing end-to-end training and gradient-based test-time adaptation that outperforms prior methods on combinatorial generalization tasks.
Title resolution pending
28 Pith papers cite this work. Polarity classification is still indexing.
28
Pith papers citing it
citation-role summary
background 3
method 1
citation-polarity summary
representative citing papers
PWO is a trust-region optimizer for autoregressive NQS that improves stability over Adam and stochastic reconfiguration methods while scaling to 1.5B-parameter models on spin systems.
JanusPipe introduces SymFold and WaveK to enable efficient 3D-parallel training for conservative MLIPs, reporting 1.51x and 1.45x average throughput gains over 1F1B and Hanayo baselines on 32 GPUs.
A hypernetwork maps style motion embeddings to LoRA updates that stylize text-driven motion diffusion models with improved generalization to unseen styles via contrastive structuring of the style space.
SEMIR replaces dense voxel computation with a learned topology-preserving graph minor that supports exact decoding and GNN-based inference for small-structure segmentation in large medical images.
CAWI replaces standard random initialization of input-to-hidden weights in randomized neural networks with samples drawn from a data-fitted copula that preserves observed feature dependencies, yielding consistent accuracy gains on 83 classification benchmarks.
Muon succeeds by guaranteeing local step-size optimality rather than by tracking any ideal global geometry, as random-spectrum and quasi-norm variants match its performance on language models.
ALU uses public data to suppress unlearning cost quadratically while characterizing distribution mismatch effects, enabling mass unlearning with maintained utility.
A neurosymbolic model augments Swin Transformers with focal sets and fuzzy logic to produce calibrated hierarchical image classifications that respect logical constraints.
mPL measures attacker-aligned privacy leakage from joint data releases and AmPL provides an adaptive way to bound it with low utility cost in ML settings.
LeGS turns density control in 3D Gaussian Splatting into a learnable RL policy whose reward is derived from a closed-form sensitivity analysis that measures each Gaussian's marginal contribution to reconstruction quality.
Preconditioned delta-rule models with a diagonal curvature approximation improve upon standard DeltaNet, GDN, and KDA by better approximating the test-time regression objective.
DUSG-Tomo-Net performs super-resolved gridless TomoSAR inversion by learning a Toeplitz-structured covariance representation from single-look nonuniform-baseline data via deep unfolding and projection enforcement.
A learnable nonlocal block that mimics classical neighbor matching and collaborative filtering on multiscale features produces competitive RAW denoising with far fewer parameters than current deep models and generalizes across sensors.
SWE-RL uses RL on software evolution data to train LLMs achieving 41% on SWE-bench Verified with generalization to other reasoning tasks.
SOAP and SOAP-Muon optimizers deliver faster convergence and higher final accuracy than Adam for NequIP and Allegro MLIPs, with the largest gains under partial force supervision.
LLM embeddings condition a generative transformer to enable faster convergence, better performance, and generalization to unseen LHC processes using a single model.
HORST uses non-commutative operator composition and a hyperbolic mirror map to combine stability from adaptive optimizers with L1 sparsity bias, outperforming AdamW across sparsity levels on vision and language tasks.
LLMs represent semantic relations geometrically via embedding distance and direction; a linear Polar Probe decodes these structures from middle-layer activations and generalizes to new entities.
A noisy quadratic system predicts large model test losses from N, B, K and outperforms Chinchilla's model for extrapolation up to 1000x compute.
AdamO modifies Adam with an orthogonality correction to ensure the spectral radius of the TD update operator stays below one, providing a theoretical stability guarantee for offline RL.
Fine-tuning LLMs on Arabic yields similar zero-shot gains on Semitic and non-Semitic languages, with chain-of-thought reasoning producing parallel benefits, indicating task alignment drives transfer more than language relatedness.
COMPASS uses semantic clustering on multilingual embeddings to select auxiliary data for PEFT adapters, outperforming linguistic-similarity baselines on multilingual benchmarks while supporting continual adaptation.
LASER couples a latent world model with a GRPO-trained RL policy to adaptively reposition sensors for sparse continuum field reconstruction, outperforming fixed and offline-optimized placement strategies across PDE and real-world benchmarks.
citing papers explorer
-
Gradient-Based Program Synthesis with Neurally Interpreted Languages
NLI autonomously discovers a vocabulary of primitive operations and interprets variable-length programs via a neural executor, allowing end-to-end training and gradient-based test-time adaptation that outperforms prior methods on combinatorial generalization tasks.
-
One More Time: Revisiting Neural Quantum States from a Reinforcement Learning Perspective
PWO is a trust-region optimizer for autoregressive NQS that improves stability over Adam and stochastic reconfiguration methods while scaling to 1.5B-parameter models on spin systems.
-
JanusPipe: Efficient Pipeline Parallel Training for Machine Learning Interatomic Potentials
JanusPipe introduces SymFold and WaveK to enable efficient 3D-parallel training for conservative MLIPs, reporting 1.51x and 1.45x average throughput gains over 1F1B and Hanayo baselines on 32 GPUs.
-
Stylized Text-to-Motion Generation via Hypernetwork-Driven Low-Rank Adaptation
A hypernetwork maps style motion embeddings to LoRA updates that stylize text-driven motion diffusion models with improved generalization to unseen styles via contrastive structuring of the style space.
-
SEMIR: Semantic Minor-Induced Representation Learning on Graphs for Visual Segmentation
SEMIR replaces dense voxel computation with a learned topology-preserving graph minor that supports exact decoding and GNN-based inference for small-structure segmentation in large medical images.
-
CAWI: Copula-Aligned Weight Initialization for Randomized Neural Networks
CAWI replaces standard random initialization of input-to-hidden weights in randomized neural networks with samples drawn from a data-fitted copula that preserves observed feature dependencies, yielding consistent accuracy gains on 83 classification benchmarks.
-
Muon is Not That Special: Random or Inverted Spectra Work Just as Well
Muon succeeds by guaranteeing local step-size optimality rather than by tracking any ideal global geometry, as random-spectrum and quasi-norm variants match its performance on language models.
-
Unlearning with Asymmetric Sources: Improved Unlearning-Utility Trade-off with Public Data
ALU uses public data to suppress unlearning cost quadratically while characterizing distribution mismatch effects, enabling mass unlearning with maintained utility.
-
A neurosymbolic Approach with Epistemic Deep Learning for Hierarchical Image Classification
A neurosymbolic model augments Swin Transformers with focal sets and fuzzy logic to produce calibrated hierarchical image classifications that respect logical constraints.
-
Metric-Normalized Posterior Leakage (mPL): Attacker-Aligned Privacy for Joint Consumption
mPL measures attacker-aligned privacy leakage from joint data releases and AmPL provides an adaptive way to bound it with low utility cost in ML settings.
-
Beyond Heuristics: Learnable Density Control for 3D Gaussian Splatting
LeGS turns density control in 3D Gaussian Splatting into a learnable RL policy whose reward is derived from a closed-form sensitivity analysis that measures each Gaussian's marginal contribution to reconstruction quality.
-
Preconditioned DeltaNet: Curvature-aware Sequence Modeling for Linear Recurrences
Preconditioned delta-rule models with a diagonal curvature approximation improve upon standard DeltaNet, GDN, and KDA by better approximating the test-time regression objective.
-
DUSG-Tomo-Net: A Deep Unfolded Neural Network for Super-Resolving Gridless Spaceborne SAR Tomography via Learned Toeplitz-Structured Covariance Representation
DUSG-Tomo-Net performs super-resolved gridless TomoSAR inversion by learning a Toeplitz-structured covariance representation from single-look nonuniform-baseline data via deep unfolding and projection enforcement.
-
Learned Nonlocal Feature Matching and Filtering for RAW Image Denoising
A learnable nonlocal block that mimics classical neighbor matching and collaborative filtering on multiscale features produces competitive RAW denoising with far fewer parameters than current deep models and generalizes across sensors.
-
SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution
SWE-RL uses RL on software evolution data to train LLMs achieving 41% on SWE-bench Verified with generalization to other reasoning tasks.
-
Beyond Adam: SOAP and Muon for Faster, Label-Efficient Training of Machine Learning Interatomic Potentials
SOAP and SOAP-Muon optimizers deliver faster convergence and higher final accuracy than Adam for NequIP and Allegro MLIPs, with the largest gains under partial force supervision.
-
One Generator, Any Process: LLM-Conditioning for the LHC
LLM embeddings condition a generative transformer to enable faster convergence, better performance, and generalization to unseen LHC processes using a single model.
-
HORST: Composing Optimizer Geometries for Sparse Transformer Training
HORST uses non-commutative operator composition and a hyperbolic mirror map to combine stability from adaptive optimizers with L1 sparsity bias, outperforming AdamW across sparsity levels on vision and language tasks.
-
Polar probe linearly decodes semantic structures from LLMs
LLMs represent semantic relations geometrically via embedding distance and direction; a linear Polar Probe decodes these structures from middle-layer activations and generalizes to new entities.
-
Predicting Large Model Test Losses with a Noisy Quadratic System
A noisy quadratic system predicts large model test losses from N, B, K and outperforms Chinchilla's model for extrapolation up to 1000x compute.
-
AdamO: A Collapse-Suppressed Optimizer for Offline RL
AdamO modifies Adam with an orthogonality correction to ensure the spectral radius of the TD update operator stays below one, providing a theoretical stability guarantee for offline RL.
-
Disentangling Linguistic Relatedness from Task Alignment in Cross-Lingual Transfer
Fine-tuning LLMs on Arabic yields similar zero-shot gains on Semitic and non-Semitic languages, with chain-of-thought reasoning producing parallel benefits, indicating task alignment drives transfer more than language relatedness.
-
COMPASS: COntinual Multilingual PEFT with Adaptive Semantic Sampling
COMPASS uses semantic clustering on multilingual embeddings to select auxiliary data for PEFT adapters, outperforming linguistic-similarity baselines on multilingual benchmarks while supporting continual adaptation.
-
LASER: Learning Active Sensing for Continuum Field Reconstruction
LASER couples a latent world model with a GRPO-trained RL policy to adaptively reposition sensors for sparse continuum field reconstruction, outperforming fixed and offline-optimized placement strategies across PDE and real-world benchmarks.
-
Learning to Segment using Summary Statistics and Weak Supervision
A novel loss function enables effective segmentation training from summary statistics combined with minimal weak pixel supervision, outperforming statistics alone on medical imaging tasks.
-
Mesh Based Simulations with Spatial and Temporal awareness
A unified training framework for mesh-based ML surrogates in CFD improves accuracy and long-horizon stability by enforcing spatial derivative consistency via multi-node prediction, using temporal cross-attention correction, and adding 3D rotary positional embeddings.
-
Photometric Super-Resolution for Improving Galaxy Morphological Measurements using Conditional Generative Adversarial Networks
Neo, a cGAN, super-resolves HSC images to HST-like quality and improves galaxy morphological parameter accuracy by factors of 2-10.
-
Probabilistic Classification and Uncertainty Quantification of Sahara Desert Climate Using Feedforward Neural Networks
A feedforward neural network delivers probabilistic climate classifications and uncertainty estimates for the Sahara Desert from 1960-1989 data, tracking temporal shifts via fluctuation analysis.