R
Published on

ICLR 2026 — Other / Unclassified

Other / Unclassified

1101 papers (0 oral)

Information Shapes Koopman Representation

On The Surprising Effectiveness of a Single Global Merging in Decentralized Learning

On the Wasserstein Geodesic Principal Component Analysis of probability measures

RAIN-Merging: A Gradient-Free Method to Enhance Instruction Following in Large Reasoning Models with Preserved Thinking Format

Gaussian certified unlearning in high dimensions: A hypothesis testing approach

Distributional Equivalence in Linear Non-Gaussian Latent-Variable Cyclic Causal Models: Characterization and Learning

Veritas: Generalizable Deepfake Detection via Pattern-Aware Reasoning

Differentially Private Domain Discovery

Monocular Normal Estimation via Shading Sequence Estimation

Causal Structure Learning in Hawkes Processes with Complex Latent Confounder Networks

From Markov to Laplace: How Mamba In-Context Learns Markov Chains

Reasoning with Sampling: Your Base Model is Smarter Than You Think

Radiometrically Consistent Gaussian Surfels for Inverse Rendering

A Representer Theorem for Hawkes Processes via Penalized Least Squares Minimization

True Self-Supervised Novel View Synthesis is Transferable

Q-RAG: Long Context Multi‑Step Retrieval via Value‑Based Embedder Training

Taming Momentum: Rethinking Optimizer States Through Low-Rank Approximation

The Polar Express: Optimal Matrix Sign Methods and their Application to the Muon Algorithm

Temporal superposition and feature geometry of RNNs under memory demands

Scaling Laws and Spectra of Shallow Neural Networks in the Feature Learning Regime

It's All Just Vectorization: einx, a Universal Notation for Tensor Operations

In-Place Test-Time Training

OpenThoughts: Data Recipes for Reasoning Models

Hallucination Begins Where Saliency Drops

Coupling Experts and Routers in Mixture-of-Experts via an Auxiliary Loss

Mean Flow Policy with Instantaneous Velocity Constraint for One-step Action Generation

BioX-Bridge: Model Bridging for Unsupervised Cross-Modal Knowledge Transfer across Biosignals

Difficult Examples Hurt Unsupervised Contrastive Learning: A Theoretical Perspective

Instilling an Active Mind in Avatars via Cognitive Simulation

Characterizing the Discrete Geometry of ReLU Networks

WAFT: Warping-Alone Field Transforms for Optical Flow

InfoNCE Induces Gaussian Distribution

Overparametrization bends the landscape: BBP transitions at initialization in simple Neural Networks

Overcoming Joint Intractability with Lossless Hierarchical Speculative Decoding

AnyUp: Universal Feature Upsampling

Addressing divergent representations from causal interventions on neural networks

Generating metamers of human scene understanding

FIRE: Frobenius-Isometry Reinitialization for Balancing the Stability–Plasticity Tradeoff

Why DPO is a Misspecified Estimator and How to Fix It

Online Learning and Equilibrium Computation with Ranking Feedback

Non-Asymptotic Analysis of (Sticky) Track-and-Stop

Pinet: Optimizing hard-constrained neural networks with orthogonal projection layers

FALCON: Few-step Accurate Likelihoods for Continuous Flows

Learning to Segment for Vehicle Routing Problems

Latent Fourier Transform

AdaSpec: Adaptive Spectrum for Enhanced Node Distinguishability

14. TIPO: Text to Image with Text Pre-sampling for Prompt Optimization

  • Topics: Optimization & Training Methods, Computer Vision

Premise Selection for a Lean Hammer

24. Mechanistic Independence: A Principle for Identifiable Disentangled Representations

  • Topics: Interpretability & Mechanistic Interpretability

Online Prediction of Stochastic Sequences with High Probability Regret Bounds

26. UniCalli: A Unified Diffusion Framework for Column-Level Generation and Recognition of Chinese Calligraphy

  • Topics: Diffusion Models & Generative AI, Graph Neural Networks, Graphs & Combinatorial

Coarse-to-Fine Learning of Dynamic Causal Structures

34. TrojanTO: Action-Level Backdoor Attacks Against Trajectory Optimization Models

  • Topics: Optimization & Training Methods, Trust & Safety

Towards Sampling Data Structures for Tensor Products in Turnstile Streams

36. Nearly Space-Optimal Graph and Hypergraph Sparsification in Insertion-Only Data Streams

  • Topics: Reinforcement Learning, Graph Neural Networks, Graphs & Combinatorial, Theory & Deep Learning Theory

Matching without Group Barrier for Heterogeneous Treatment Effect Estimation

Conditional Independent Component Analysis for Estimating Causal Structure with Latent Variables

Boosting Open Set Recognition Performance through Modulated Representation Learning

Adaptive Width Neural Networks

TRIDENT: Cross-Domain Trajectory Spatio-Temporal Representation via Distance-Preserving Triplet Learning

A Generalized Geometric Theoretical Framework of Centroid Discriminant Analysis for Linear Classification of Multi-dimensional Data

Behavior Learning (BL)

Unified In-Context Video Editing

Detect, Decide, Unlearn: A Transfer-Aware Framework for Continual Learning

Inferring the Invisible: Neuro-Symbolic Rule Discovery for Missing Value Imputation

NeoBERT: A Next Generation BERT

94. Synthetic History: Evaluating Visual Representations of the Past in Diffusion Models

  • Topics: Diffusion Models & Generative AI, Computer Vision

Gradient-Based Program Synthesis with Neurally Interpreted Languages

Mod-Adapter: Tuning-Free and Versatile Multi-concept Personalization via Modulation Adapter

Evaluating GFlowNet from partial episodes for stable and flexible policy-based training

CLEAR: Calibrated Learning for Epistemic and Aleatoric Risk

JAPAN: Joint Adaptive Prediction Areas with Normalising Flow

Latent Geometry-Driven Network Automata for Complex Network Dismantling

CoLA: Co-Calibrated Logit Adjustment for Long-Tailed Semi-Supervised Learning

Topology-Preserved Auto-regressive Mesh Generation in the Manner of Weaving Silk

Understanding the Robustness of Distributed Self-Supervised Learning Frameworks Against Non-IID Data

LumiTex: Towards High-Fidelity PBR Texture Generation with Illumination Context

DA2^{2}: Depth Anything in Any Direction

Random-projection ensemble dimension reduction

Entropic Confinement and Mode Connectivity in Overparameterized Neural Networks

LMask: Learn to Solve Constrained Routing Problems with Lazy Masking

Deft Scheduling of Dynamic Cloud Workflows with Varying Deadlines via Mixture-of-Experts

Rethinking Radiology Report Generation: From Narrative Flow to Topic-Guided Findings

Think-While-Generating: On-the-Fly Reasoning for Personalized Long-Form Generation

Lifelong Learning with Behavior Consolidation for Vehicle Routing

Scalable and Adaptive Trust-Region Learning via Projection Convex Hull

Rex-Thinker: Grounded Object Referring via Chain-of-Thought Reasoning

OmniSTVG: Toward Spatio-Temporal Omni-Object Video Grounding

GOT-Edit: Geometry-Aware Generic Object Tracking via Online Model Editing

Midway Network: Learning Representations for Recognition and Motion from Latent Dynamics

Riemannian Zeroth-Order Gradient Estimation with Structure-Preserving Metrics for Geodesically Incomplete Manifolds

Error Feedback for Muon and Friends

Byzantine-Robust Federated Learning with Learnable Aggregation Weights

SpectralGCD: Spectral Concept Selection and Cross-modal Representation Learning for Generalized Category Discovery

CubeBench: Diagnosing Interactive, Long-Horizon Physical Intelligence under Partial Observations

Improved ℓp\ell_{p} Regression via Iteratively Reweighted Least Squares

Layerwise Federated Learning for Heterogeneous Quantum Clients using Quorus

Memento: Toward an All-Day Proactive Assistant for Ultra-Long Streaming Video

vCache: Verified Semantic Prompt Caching

CLIP Behaves like a Bag-of-Words Model Cross-modally but not Uni-modally

Lookup multivariate Kolmogorov-Arnold Networks

Test-Time Training Done Right

Remaining-data-free Machine Unlearning by Suppressing Sample Contribution

Action-Guided Attention for Video Action Anticipation

GOLDILOCS: GENERAL OBJECT-LEVEL DETECTION AND LABELING OF CHANGES IN SCENES

From Vicious to Virtuous Cycles: Synergistic Representation Learning for Unsupervised Video Object-Centric Learning

Fractional-Order Spiking Neural Network

Log-Linear Attention

Procedural Mistake Detection via Action Effect Modeling

FedOpenMatch: Towards Semi-Supervised Federated Learning in Open-Set Environments

Learning AND–OR Templates for Compositional Representation in Art and Design

TAPTRv3: Spatial and Temporal Context Foster Robust Tracking of Any Point in Long Video

Latent Stochastic Interpolants

VeriEquivBench: An Equivalence Score for Ground-Truth-Free Evaluation of Formally Verifiable Code

Free Lunch for Stabilizing Rectified Flow Inversion

MOLM: Mixture of LoRA Markers

Synthesising Counterfactual Explanations via Label-Conditional Gaussian Mixture Variational Autoencoders

Debugging Concept Bottleneck Models through Removal and Retraining

CyclicReflex: Improving Reasoning Models via Cyclical Reflection Token Scheduling

LD-MoLE: Learnable Dynamic Routing for Mixture of LoRA Experts

When Machine Learning Gets Personal: Evaluating Prediction and Explanation

Sheaves Reloaded: A Direction Awakening

Label Smoothing Improves Machine Unlearning

Fair Classification by Direct Intervention on Operating Characteristics

Cooperative Sheaf Neural Networks

Noise Tolerance of Distributionally Robust Learning

Conditioned Initialization for Attention

From Neural Networks to Logical Theories: The Correspondence between Fibring Modal Logics and Fibring Neural Networks

The Serial Scaling Hypothesis

Fisher-Rao Sensitivity for Out-of-Distribution Detection in Deep Neural Networks

Towards Safe Reasoning in Large Reasoning Models via Corrective Intervention

LDT: Layer-Decomposition Training Makes Networks More Generalizable

Stable and Scalable Deep Predictive Coding Networks with Meta-Prediction Errors

Nonparametric Contextual Online Bilateral Trade

Deep-ICE: The first globally optimal algorithm for empirical risk minimization of two-layer maxout and ReLU networks

Bayesian Influence Functions for Hessian-Free Data Attribution

Finite-Time Analysis of Actor-Critic Methods with Deep Neural Network Approximation

Learning the Inverse Temperature of Ising Models under Hard Constraints using One Sample

A Recovery Guarantee for Sparse Neural Networks

Online Rounding and Learning Augmented Algorithms for Facility Location

Test-Time Adaptation without Source Data for Out-of-Domain Bioactivity Prediction

Best-of-N through the Smoothing Lens: KL Divergence and Regret Analysis

Bayesian Post Training Enhancement of Regression Models with Calibrated Rankings

TianQuan-S2S: A Subseasonal-to-Seasonal Global Weather Model via Incorporate Climatology State

ROC-n-reroll: How verifier imperfection affects test-time scaling

Proximal Supervised Fine-Tuning

A New Paradigm for Genome-wide DNA Methylation Prediction Without Methylation Input

CORE: Concept-Oriented Reinforcement for Bridging the Definition–Application Gap in Mathematical Reasoning

Nef-Net v2: Adapting Electrocardio Panorama in the wild

ProstaTD: Bridging Surgical Triplet from Classification to Fully Supervised Detection

Translating Flow to Policy via Hindsight Online Imitation

ExGRPO: Learning to Reason from Experience

Latent Adaptation of Foundation Policies for Sim-to-Real Transfer

A General Spatio-Temporal Backbone with Scalable Contextual Pattern Bank for Urban Continual Forecasting

Lipschitz Bandits with Stochastic Delayed Feedback

PGRF-Net: A Prototype-Guided Relational Fusion Network for Diagnostic Multivariate Time-Series Anomaly Detection

EMFuse: Energy-based Model Fusion for Decision Making

Rethinking Reasoning in Document Ranking: Why Chain-of-Thought Falls Short

Hybrid Deep Searcher: Scalable Parallel and Sequential Search Reasoning

Harder Is Better: Boosting Mathematical Reasoning via Difficulty-Aware GRPO and Multi-Aspect Question Reformulation

The Potential of CoT for Reasoning: A Closer Look at Trace Dynamics

Tools are under-documented: Simple Document Expansion Boosts Tool Retrieval

Toward Complex-Valued Neural Networks for Waveform Generation

MENLO: From Preferences to Proficiency – Evaluating and Modeling Native-like Quality Across 47 Languages

End-to-end Listen, Look, Speak and Act

Long Chain-of-Thought Reasoning Across Languages

Beyond Grid-Locked Voxels: Neural Response Functions for Continuous Brain Encoding

An Information-Theoretic Framework For Optimizing Experimental Design To Distinguish Probabilistic Neural Codes

Understanding Task Vectors in In-Context Learning: Emergence, Functionality, and Limitations

A Brain-Inspired Gating Mechanism Unlocks Robust Computation in Spiking Neural Networks

Tokenizing Single-Channel EEG with Time-Frequency Motif Learning

Learning Mixtures of Linear Dynamical Systems via Hybrid Tensor-EM Method

AutoMetrics: Approximate Human Judgments with Automatically Generated Evaluators

LipNeXt: Scaling up Lipschitz-based Certified Robustness to Billion-parameter Models

QLCoder: A Query Synthesizer For Static Analysis of Security Vulnerabilities

Mix-Ecom: Towards Mixed-Type E-Commerce Dialogues with Complex Domain Rules

Better Bounds for the Distributed Experts Problem

SONA: Learning Conditional, Unconditional, and Matching-Aware Discriminator

Diverse Dictionary Learning

UniVideo: Unified Understanding, Generation, and Editing for Videos

TrajTok: What makes for a good trajectory tokenizer in behavior generation?

Any-step Generation via N-th Order Recursive Consistent Velocity Field Estimation

Scalable Chain of Thoughts via Elastic Reasoning

Learning Deformable Body Interactions With Adaptive Spatial Tokenization

890. Discrete Audio Tokens: More Than a Survey!

  • Topics: Audio & Speech

891. Learning Posterior Predictive Distributions for Node Classification from Synthetic Graph Priors

  • Topics: Graph Neural Networks, Graphs & Combinatorial

Influence without Confounding: Causal Discovery from Temporal Data with Long-term Carry-over Effects

Captain Cinema: Towards Short Movie Generation

DreamSwapV: Mask-guided Subject Swapping for Any Customized Video Editing

Journey to the Centre of Cluster: Harnessing Interior Nodes for A/B Testing under Network Interference

ContextGen: Contextual Layout Anchoring for Identity-Consistent Multi-Instance Generation

Off-Policy Evaluation for Ranking Policies under Deterministic Logging Policies

Toward Enhancing Representation Learning in Federated Multi-Task Settings

Maximizing Incremental Information Entropy for Contrastive Learning

EgoTwin: Dreaming Body and View in First Person

Proper Velocity Neural Networks

ReFocusEraser: Refocusing for Small Object Removal with Robust Context-Shadow Repair

CORDS - Continuous Representations of Discrete Structures

Closing the Modality Gap Aligns Group-Wise Semantics

LVTINO: LAtent Video consisTency INverse sOlver for High Definition Video Restoration

Robust Selective Activation with Randomized Temporal K-Winner-Take-All in Spiking Neural Networks for Continual Learning

Human3R: Everyone Everywhere All at Once

UP2You: Fast Reconstruction of Yourself from Unconstrained Photo Collections

Beyond Student: An Asymmetric Network for Neural Network Inheritance

DiMeR: Disentangled Mesh Reconstruction Model with Normal-only Geometry Training

Horseshoe Splatting: Handling Structural Sparsity for Uncertainty-Aware Gaussian-Splatting Radiance Field Rendering

IDER: IDempotent Experience Replay for Reliable Continual Learning

Large Depth Completion Model from Sparse Observations

Rethinking Continual Learning with Progressive Neural Collapse

Universal Beta Splatting

Learning Survival Distributions with Individually Calibrated Asymmetric Laplace Distribution

Sparkle: A Robust and Versatile Representation for Point Cloud-based Human Motion Capture

Accelerated Parallel Tempering via Neural Transports

Internal Evaluation of Density-Based Clusterings with Noise

Initialization Schemes for Kolmogorov–Arnold Networks: An Empirical Study

Solving the 2-norm k-hyperplane clustering problem via multi-norm formulations

Hystar: Hypernetwork-driven Style-adaptive Retrieval via Dynamic SVD Modulation

Deep FlexQP: Accelerated Nonlinear Programming via Deep Unfolding

Revisiting Sharpness-Aware Minimization: A More Faithful and Effective Implementation

Chain-of-Context Learning: Dynamic Constraint Understanding for Multi-Task VRPs

SONIC: Spectral Oriented Neural Invariant Convolutions

On the Benefits of Weight Normalization for Overparameterized Matrix Sensing

Fantastic Tractor-Dogs and How Not to Find Them With Open-Vocabulary Detectors

The Power of Small Initialization in Noisy Low-Tubal-Rank Tensor Recovery

A Physics-Inspired Optimizer: Velocity Regularized Adam

Online Minimization of Polarization and Disagreement via Low-Rank Matrix Bandits

Learning Distributions over Permutations and Rankings with Factorized Representations

Learning Boltzmann Generators via Constrained Mass Transport

No outlier channels but with outlier blocks

Muon Outperforms Adam in Tail-End Associative Memory Learning

MILPnet: A Multi-Scale Architecture with Geometric Feature Sequence Representations for Advancing MILP Problems

Sublinear Time Quantum Algorithm for Attention Approximation

Cautious Optimizers: Improving Training with One Line of Code

STDDN: A Physics-Guided Deep Learning Framework for Crowd Simulation

CroCoDiLight: Repurposing Cross-View Completion Encoders for Relighting

Cactus: Accelerating Auto-Regressive Decoding with Constrained Acceptance Speculative Sampling

Deep Global-sense Hard-negative Discriminative Generation Hashing for Cross-modal Retrieval

ReactDance: Hierarchical Representation for High-Fidelity and Coherent Long-Form Reactive Dance Generation

SCRAPL: Scattering Transform with Random Paths for Machine Learning

Seeing What’s Not There: Negation Understanding Needs More Than Training

TD-MoE: Tensor Decomposition for MoE Models

Leveraging Data to Say No: Memory Augmented Plug-and-Play Selective Prediction

vAttention: Verified Sparse Attention via Sampling

Free Energy Mixer

Group Representational Position Encoding

GUIDE: Gated Uncertainty-Informed Disentangled Experts for Long-tailed Recognition

Short Window Attention Enables Long-Term Memorization

Adaptive Gaussian Expansion for On-the-fly Category Discovery

Nonparametric Teaching of Attention Learners

Spectral Attention Steering for Prompt Highlighting

Go Beyond Earth: Understanding Human Actions and Scenes in Microgravity Environments

QUEST: A robust attention formulation using query-modulated spherical attention

MoM: Linear Sequence Modeling with Mixture-of-Memories

The Quest for Generalizable Motion Generation: Data, Model, and Evaluation

InfoBridge: Mutual Information estimation via Bridge Matching

Discrete Variational Autoencoding via Policy Search

Bi-Lipschitz Autoencoder With Injectivity Guarantee

Toward Effective Tool-Integrated Reasoning via Self-Evolved Preference Learning

Conformal Prediction with Corrupted Labels: Uncertain Imputation and Robust Re-weighting

AlphaFlow: Understanding and Improving MeanFlow Models

Trust The Typical

Attention Smoothing Is All You Need For Unlearning

InfBaGel: Human-Object-Scene Interaction Generation with Dynamic Perception and Iterative Refinement

Decoupling the Class Label and the Target Concept in Machine Unlearning

Enhancing Hallucination Detection through Noise Injection

Discern Truth from Falsehood: Reducing Over-Refusal via Contrastive Refinement

Unlearning Evaluation through Subset Statistical Independence

Attribution-Guided Decoding

When Scores Learn Geometry: Rate Separations under the Manifold Hypothesis

VeriTrail: Closed-Domain Hallucination Detection with Traceability

TreeGrad-Ranker: Feature Ranking via O(L)O(L)-Time Gradients for Decision Trees

On Universality of Deep Equivariant Networks

RedacBench: Can AI Erase Your Secrets?

On the Lipschitz Continuity of Set Aggregation Functions and Neural Networks for Sets

Time Is All It Takes: Spike-Retiming Attacks on Event-Driven Spiking Neural Networks

Enhancing Learning with Noisy Labels via Rockafellian Relaxation

How to Cure Newton for Unlearning Neural Networks? An Empirical Study from the Hessian Perspective

Statistical Advantage of Softmax Attention: Insights from Single-Location Regression

Fast Catch-Up, Late Switching: Optimal Batch Size Scheduling via Functional Scaling Laws

Differentially Private Equilibrium Finding in Polymatrix Games

XIL: Cross-Expanding Incremental Learning

IMSE: Intrinsic Mixture of Spectral Experts Fine-tuning for Test-Time Adaptation

Bayesian Evidence-Driven Prototype Evolution for Federated Domain Adaptation

Think in Parallel, Answer as One: Logit Averaging for Open-Ended Reasoning

Understanding the Emergence of Seemingly Useless Features in Next-Token Predictors

Dynamic Chunking for End-to-End Hierarchical Sequence Modeling

A Faster Parameter-Free Regret Matching Algorithm

Aligning Deep Implicit Preferences by Learning to Reason Defensively

Contextual Multi-Armed Bandits with Minimum Aggregated Revenue Constraints

Near Optimal Robust Federated Learning Against Data Poisoning Attack

Heads collapse, features stay: Why Replay needs big buffers

High-Dimensional Analysis of Single-Layer Attention for Sparse-Token Classification

Feedback-driven recurrent quantum neural network universality

t-SNE Exaggerates Clusters, Provably

To Augment or Not to Augment? Diagnosing Distributional Symmetry Breaking

Long-Document QA with Chain-of-Structured-Thought and Fine-Tuned SLMs

Characterizing Pattern Matching and Its Limits on Compositional Task Structures

Detecting Invariant Manifolds in ReLU-Based RNNs

Extreme Weather Nowcasting via Local Precipitation Pattern Prediction

Glance and Focus Reinforcement for Pan-cancer Screening

Singleton-Optimized Conformal Prediction

Learning Self-Critiquing Mechanisms for Region-Guided Chest X-Ray Report Generation

Physics-Constrained Fine-Tuning of Flow-Matching Models for Generation and Inverse Problems

The False Promise of Zero-Shot Super-Resolution in Machine-Learned Operators

Bayesian Parameter Shift Rules in Variational Quantum Eigensolvers

Advancing Universal Deep Learning for Electronic-Structure Hamiltonian Prediction of Materials

Value Flows

AnyTouch 2: General Optical Tactile Representation Learning For Dynamic Tactile Perception

Safe Exploration via Policy Priors

R2PS: Worst-Case Robust Real-Time Pursuit Strategies under Partial Observability

Neural Latent Arbitrary Lagrangian-Eulerian Grids for Fluid-Solid Interaction

Revenue Maximization Under Sequential Price Competition Via The Estimation Of ss-Concave Demand Functions

Contextual Causal Bayesian Optimisation

STORM: Synergistic Cross-Scale Spatio-Temporal Modeling for Weather Forecasting

Accelerated Learning with Linear Temporal Logic using Differentiable Simulation

Demystifying Deep Search: A Holistic Evaluation with Hint-free Multi-Hop Questions and Factorised Metrics

Tina: Tiny Reasoning Models via LoRA

Log-Augmented Generation: Scaling Test-Time Reasoning with Reusable Computation

Stacked from One: Multi-Scale Self-Injection for Context Window Extension

Strategic Scaling of Test-Time Compute: A Bandit Learning Approach

Silent Leaks: Implicit Knowledge Extraction Attack on RAG Systems

Fine-tuning Done Right in Model Editing

DirMoE: Dirichlet-Routed Mixture of Experts

TNT: Improving Chunkwise Training for Test-Time Memorization

MILCO: Learned Sparse Retrieval Across Languages via a Multilingual Connector

Tokenisation over Bounded Alphabets is Hard

Unified Analyses for Hierarchical Federated Learning: Topology Selection under Data Heterogeneity

SongEcho: Towards Cover Song Generation via Instance-Adaptive Element-wise Linear Modulation

MTVCraft: Tokenizing 4D Motion for Arbitrary Character Animation

Prior-based Noisy Text Data Filtering: Fast and Strong Alternative For Perplexity

Goedel-Prover-V2: Scaling Formal Theorem Proving with Scaffolded Data Synthesis and Self-Correction

Modeling Others' Minds as Code

Continuous multinomial logistic regression for neural decoding

Decoding Open-Ended Information Seeking Goals from Eye Movements in Reading

TRIBE: TRImodal Brain Encoder for whole-brain fMRI response prediction

Neuro-Symbolic Decoding of Neural Activity

Training Deep Normalization-Free Spiking Neural Networks with Lateral Inhibition

Read the Room: Video Social Reasoning with Mental-Physical Causal Chains

Transfer Learning in Infinite Width Feature Learning Networks

Homeostatic Adaptation of Optimal Population Codes under Metabolic Stress

Bidirectional Predictive Coding

ECHO: Toward Contextual Seq2Seq Paradigms in Large EEG Models

ArtUV: Artist-style UV Unwrapping

InputDSA: Demixing, then comparing recurrent and externally driven dynamics

FastVMT: Eliminating Redundancy in Video Motion Transfer

MobileKGQA: On-Device KGQA System on Dynamic Mobile Environments

AVEX: What Matters for Animal Vocalization Encoding

EXP-Bench: Can AI Conduct AI Research Experiments?

Prior-aware and Context-guided Group Sampling for Active Probabilistic Subsampling

Exposing and Defending the Achilles' Heel of Video Mixture-of-Experts

Resisting Contextual Interference in RAG via Parametric-Knowledge Reinforcement

Your Models Have Thought Enough: Training Large Reasoning Models to Stop Overthinking

Decoupled DMD: CFG Augmentation as the Spear, Distribution Matching as the Shield

Conformalized Hierarchical Calibration for Uncertainty-Aware Adaptive Hashing

DeAltHDR: Learning HDR Video Reconstruction from Degraded Alternating Exposure Sequences

Less Gaussians, Texture More: 4K Feed-Forward Textured Splatting

SWERank: Software Issue Localization with Code Ranking

Taming Imperfect Process Verifiers: A Sampling Perspective on Backtracking

Depth Anything with Any Prior

Distribution-informed Online Conformal Prediction

DiffTrans: Differentiable Geometry-Materials Decomposition for Reconstructing Transparent Objects

Consistent Low-Rank Approximation

Learning to Reason for Hallucination Span Detection

Paper Copilot: Tracking the Evolution of Peer Review in AI Conferences

Towards Safe and Optimal Online Bidding: A Modular Look-ahead Lyapunov Framework

Structure Learning from Time-Series Data with Lag-Agnostic Structural Prior

Statistical and structural identifiability in representation learning

On Measuring Influence in Avoiding Undesired Future

Direct Doubly Robust Estimation of Conditional Quantile Contrasts

FACM: Flow-Anchored Consistency Models

Lossy Common Information in a Learnable Gray-Wyner Network

Implicit Inversion turns CLIP into a Decoder

Physically-Guided Optical Inversion Enable Non-Contact Side-Channel Attack on Isolated Screens

Why Prototypes Collapse: Diagnosing and Preventing Partial Collapse in Prototypical Self-Supervised Learning

Multilingual Routing in Mixture-of-Experts

FreeViS: Training-free Video Stylization with Inconsistent References

Pixel to Gaussian: Ultra-Fast Continuous Super-Resolution with 2D Gaussian Modeling

Polynomial, trigonometric, and tropical activations

LORE: Jointly Learning The Intrinsic Dimensionality and Relative Similarity Structure from Ordinal Data

Unsupervised Representation Learning - an Invariant Risk Minimization Perspective

Learning Physics-Grounded 4D Dynamics with Neural Gaussian Force Fields

Beyond Skeletons: Learning Animation Directly from Driving Videos with Same2X Training Strategy

Contrastive Predictive Coding Done Right for Mutual Information Estimation

Learning Explicit Single-Cell Dynamics Using ODE Representations

DeepFRC: An End-to-End Deep Learning Model for Functional Registration and Classification

Learning a distance measure from the information-estimation geometry of data

The Lie of the Average: How Class Incremental Learning Evaluation Deceives You?

Sharp Monocular View Synthesis in Less Than a Second

Splat the Net: Radiance Fields with Splattable Neural Primitives

Navigating the Accuracy-Size Trade-Off with Flexible Model Merging

TRACED: Transition-aware Regret Approximation with Co-learnability for Environment Design

Fused-Planes: Why Train a Thousand Tri-Planes When You Can Share?

One-Prompt Strikes Back: Sparse Mixture of Experts for Prompt-based Continual Learning

LiTo: Surface Light Field Tokenization

HDR-NSFF: High Dynamic Range Neural Scene Flow Fields

Splat Feature Solver

DiffBED: Scaling Bayesian Experimental Design to High-Dimensions

Neural Posterior Estimation with Latent Basis Expansions

Conformal Prediction for Long-Tailed Classification

Multi-Object System Identification from Videos

When Shift Happens - Confounding Is to Blame

SONATA: Synergistic Coreset Informed Adaptive Temporal Tensor Factorization

Sharing State Between Prompts and Programs

Generalizing Linear Autoencoder Recommenders with Decoupled Expected Quadratic Loss

RADAR: Learning to Route with Asymmetry-aware Distance Representations

Combination-of-Experts with Knowledge Sharing for Cross-Task Vehicle Routing Problems

Learning to Solve Orienteering Problem with Time Windows and Variable Profits

Celo: Training Versatile Learned Optimizers on a Compute Diet

1984. UI-Ins: Enhancing GUI Grounding with Multi-Perspective Instruction as Reasoning

  • Topics: Other / Unclassified

Hinge Regression Tree: A Newton Method for Oblique Regression Tree Splitting

On Smoothness Bounds for Non-Clairvoyant Scheduling with Predictions

Submodular Function Minimization with Dueling Oracle

RepSpec: Structural Re-parameterized Draft Model Training for Speculative Decoding

Plan then Act: Bi-level CAD Command Sequence Generation

Sapiens2

MnemoDyn: Learning Resting State Dynamics from 4040K FMRI sequences

Iterative Training of Physics-Informed Neural Networks with Fourier-enhanced Features

Stable-LoRA: Stabilizing Feature Learning of Low-Rank Adaptation

DiaBlo: Diagonal Blocks Are Sufficient For Finetuning

Towards Understanding The Calibration Benefits of Sharpness-Aware Minimization

Imagine How To Change: Explicit Procedure Modeling for Change Captioning

Toward Principled Flexible Scaling for Self-Gated Neural Activation

UNITE: Universal kNowledge Integration from Task-specific Experts

LANE: Label-Aware Noise Elimination for Fine-Grained Text Classification

Intrinsic Lorentz Neural Network

Matting Anything 2: Towards Video Matting for Anything

(U)NFV: (Un)Supervised Neural Finite Volume Methods for Solving Hyperbolic PDEs

GmNet: Revisiting Gating Mechanisms From A Frequency View

Exploring Specular Reflection Inconsistency for Generalizable Face Forgery Detection

On learning linear dynamical systems in context with attention layers

Enabling Your Forensic Detector Know How Well It Performs on Distorted Samples

Why Attention Patterns Exist: A Unifying Temporal Perspective Analysis

Scaling Attention via Feature Sparsity

Salient Object Ranking via Cyclical Perception-Viewing Interaction Modeling

FASA: FREQUENCY-AWARE SPARSE ATTENTION

Encoder-only Next Token Prediction

2104. Black-Box Privacy Attacks on Shared Representations in Multitask Learning

  • Topics: Trust & Safety

Equivariant Splitting: Self-supervised learning from incomplete data

Generalized Parallel Scaling with Interdependent Generations

Federated Learning with Profile Mapping under Distribution Shifts and Drifts

STAT: Skill-Targeted Adaptive Training

Optimal Transport-Induced Samples against Out-of-Distribution Overconfidence

Complementing Self-Consistency with Cross-Model Disagreement for Uncertainty Quantification

Revisiting the Past: Data Unlearning with Model State History

SNAP-UQ: Self-supervised Next-Activation Prediction for Single-Pass Uncertainty in TinyML

Beyond In-Domain Detection: SpikeScore for Cross-Domain Hallucination Detection

A Unified Total Variation Framework for Membrane Potential Perturbation Dynamic

Flower: A Flow-Matching Solver for Inverse Problems

EditLens: Quantifying the Extent of AI Editing in Text

SESaMo: Symmetry-Enforcing Stochastic Modulation for Normalizing Flows

Bayesian Neural Networks for Functional ANOVA Model

Robust Generalized Schrödinger Bridge via Sparse Variational Gaussian Processes

SFBD-OMNI: Bridge models for lossy measurement restoration with limited clean samples

Towards Understanding the Nature of Attention with Low-Rank Sparse Decomposition

Adaptive Canonicalization with Application to Invariant Anisotropic Geometric Networks

Statistical Guarantees in the Search for Less Discriminatory Algorithms

BARREL: Boundary-Aware Reasoning for Factual and Reliable LRMs

Dual Randomized Smoothing: Beyond Global Noise Variance

Untraceable DeepFakes via Traceable Fingerprint Elimination

LitmusValues: Will AI Tell Lies to Save Sick Children? Litmus-Testing AI Values Prioritization with AIRiskDilemmas

Certifying the Full YOLO Pipeline: A Probabilistic Verification Approach

Deterministic Bounds and Random Estimates of Metric Tensors on Neuromanifolds

Distributionally Robust Classification for Multi-source Unsupervised Domain Adaptation

TiTok: Transfer Token-level Knowledge via Contrastive Excess to Transplant LoRA

Solving Football by Exploiting Equilibrium Structure of 2p0s Differential Games with One-Sided Information

Infinite Horizon Markov Economies

Does the Data Processing Inequality Reflect Practice? On the Utility of Low-Level Tasks

Scaling Laws of SignSGD in Linear Regression: When Does It Outperform SGD?

Dimension-Free Decision Calibration for Nonlinear Loss Functions

Diversified Multinomial Logit Contextual Bandits

Why Ask One When You Can Ask kk? Learning-to-Defer to the Top-kk Experts

FAME: Formal Abstract Minimal Explanation for Neural Networks

Tuning the burn-in phase in training recurrent neural networks improves their performance

SHE-LoRA: Selective Homomorphic Encryption for Federated Tuning with Heterogeneous LoRA

Metric kk-clustering using only Weak Comparison Oracles

Physics-informed learning under mixing: How physical knowledge speeds up learning

Improved high-dimensional estimation with Langevin dynamics and stochastic weight averaging

High-dimensional limit theorems for SGD: Momentum and Adaptive Step-sizes

Chessformer: A Unified Architecture for Chess Modeling

Neural Networks Learn Generic Multi-Index Models Near Information-Theoretic Limit

Price of Quality: Sufficient Conditions for Sparse Recovery using Mixed-Quality Data

A Derandomization Framework for Structure Discovery: Applications in Neural Networks and Beyond

Statistical Guarantees for Approximate Stationary Points of Shallow Neural Networks

2330. Unleashing LLMs in Bayesian Optimization: Preference-Guided Framework for Scientific Discovery

  • Topics: LLMs & Foundation Models, Optimization & Training Methods

Flow-Disentangled Feature Importance

From Data Statistics to Feature Geometry: How Correlations Shape Superposition

Better Learning-Augmented Spanning Tree Algorithms via Metric Forest Completion

Unified Brain Surface and Volume Registration

Distilling and Adapting: A Topology-Aware Framework for Zero-Shot Interaction Prediction in Multiplex Biological Networks

Relative Value Learning

Generalized Spherical Neural Operators: Green’s Function Formulation

End-to-End Probabilistic Framework for Learning with Hard Constraints

OrthoSolver: A Neural Proper Orthogonal Decomposition Solver For PDEs

Dual Goal Representations

A Spectral-Grassmann Wasserstein metric for operator representations of dynamical systems

LD-EnSF: Synergizing Latent Dynamics with Ensemble Score Filters for Fast Data Assimilation with Sparse Observations

PRISM-Physics: Causal DAG-Based Process Evaluation for Physics Reasoning

Invert4TVG: A Temporal Video Grounding Framework with Inversion Tasks Preserving Action Understanding Ability

Towards a Certificate of Trust: Task-Aware OOD Detection for Scientific AI

Tensor learning with orthogonal, Lorentz, and symplectic symmetries

Distributional value gradients for stochastic environments

Contact-guided Real2Sim from Monocular Video with Planar Scene Primitives

A Simple "Motivation" Can Enhance Reinforcement Finetuning of Large Reasoning Models

Type-Compliant Adaptation Cascades

Beyond Noisy-TVs: Noise-Robust Exploration Via Learning Progress Monitoring

Learning to Grasp Anything By Playing with Random Toys

In-Context Learning for Pure Exploration

Perturbed Dynamic Time Warping: A Probabilistic Framework and Generalized Variants

Reasoning on Time-Series for Financial Technical Analysis

Point-wise Anomaly Detection via Fold-bifurcation ODE

The Markovian Thinker: Architecture-Agnostic Linear Scaling of Reasoning

Disentangled Representation Learning for Parametric Partial Differential Equations

Codified Finite-state Machines for Role-playing

Learning Facts at Scale with Active Reading

MATHMO: Automated Mathematical Modeling Through Adaptive Search

The Imitation Game: Turing Machine Imitator is Length Generalizable Reasoner

On the Wings of Imagination: Conflicting Script-based Multi-role Framework for Humor Caption Generation

EntropyLong: Effective Long-Context Training via Predictive Uncertainty

Reducing Symmetry Increase in Equivariant Neural Networks

Convex Dominance in Deep Learning I: A Scaling Law of Loss and Learning Rate

A Dense Subset Index for Collective Query Coverage

Reverse-Engineered Reasoning for Open-Ended Generation

Segment-Level Attribution for Selective Learning of Long Reasoning Traces

Decision Aggregation under Quantal Response

Diversity-Incentivized Exploration for Versatile Reasoning

HardcoreLogic: Challenging Large Reasoning Models with Long-tail Logic Puzzle Games

Mixture of Mini Experts: Overcoming the Linear Layer Bottleneck in Multiple Instance Learning

Readout Representation: Redefining Neural Codes by Input Recovery

Mixture of Cognitive Reasoners: Modular Reasoning with Brain-Like Specialization

Difference Predictive Coding for Training Spiking Neural Networks

Discovering heterogeneous synaptic plasticity rules via large-scale neural evolution

Spike-based Digital Brain: a novel fundamental model for brain activity analysis

PredNext: Explicit Cross-View Temporal Prediction for Unsupervised Learning in Spiking Neural Networks

EgoBrain: Synergizing Minds and Eyes For Human Action Understanding

PSDNorm: Temporal Normalization for Deep Learning in Sleep Staging

ContextPRM: Leveraging Contextual Coherence for multi-domain Test-Time Scaling

Online Pseudo-Zeroth-Order Training of Neuromorphic Spiking Neural Networks

Deep Think with Confidence

Interact-RAG: Reason and Interact with the Corpus, Beyond Black-Box Retrieval

Spatial Mental Modeling from Limited Views

Adaptive Test-Time Training for Predicting Need for Invasive Mechanical Ventilation in Multi-Center Cohorts

Pushing Test-Time Scaling Limits of Deep Search with Asymmetric Verification

Gauge-invariant representation holonomy

PairFlow: Closed-Form Source-Target Coupling for Few-Step Generation in Discrete Flow Models

Speculative Speculative Decoding

Positional Encoding Field

Decoupled MeanFlow: Turning Flow Models into Flow Maps for Accelerated Sampling

Distributional Machine Unlearning via Selective Data Removal

Distributed Algorithms for Euclidean Clustering

PosterCraft: Rethinking High-Quality Aesthetic Poster Generation in a Unified Framework

Multiple Token Divergence: Measuring and Steering In-Context Computation Density

NAB: Neural Adaptive Binning for Sparse-View CT reconstruction

Unleashing Guidance Without Classifiers for Human-Object Interaction Animation

DanceTogether: Generating Interactive Multi-Person Video without Identity Drifting

Machine Unlearning under Retain–Forget Entanglement

MathFimer: Enhancing Mathematical Reasoning by Expanding Reasoning Steps through Fill-in-the-Middle Task

Embracing Discrete Search: A Reasonable Approach to Causal Structure Learning

Counterfactual Structural Causal Bandits

IGC-Net for conditional average potential outcome estimation over time

ActiveCQ: Active Estimation of Causal Quantities

Dual Perspectives on Non-Contrastive Self-Supervised Learning

Fast and Stable Riemannian Metrics on SPD Manifolds via Cholesky Product Geometry

DeepAFL: Deep Analytic Federated Learning

KeepLoRA: Continual Learning with Residual Gradient Adaptation

Scalable Multi-Task Low-Rank Model Adaptation

A Study on PAVE Specification for Learnware

Energy-Regularized Sequential Model Editing on Hyperspheres

Learning to Generate Stylized Handwritten Text via a Unified Representation of Style, Content, and Noise

Beyond Aggregation: Guiding Clients in Heterogeneous Federated Learning

Federated ADMM from Bayesian Duality

What is the Relationship between Tensor Factorizations and Circuits (and How Can We Exploit it)?

2832. DVD-Quant: Data-free Video Diffusion Transformers Quantization

  • Topics: LLMs & Foundation Models, Diffusion Models & Generative AI, Efficiency & Compression

Fast Estimation of Wasserstein Distances via Regression on Sliced Wasserstein Distances

Robustness of Probabilistic Models to Low-Quality Data: A Multi-Perspective Analysis

Know When to Abstain: Optimal Selective Classification with Likelihood Ratios

Mesh Splatting for End-to-end Multiview Surface Reconstruction

Refine Now, Query Fast: A Decoupled Refinement Paradigm for Implicit Neural Fields

A Case for Library-Level k-Means Binning in Histogram Gradient-Boosted Trees

2865. Segment Any Events with Language

  • Topics: Other / Unclassified

FMIP: Joint Continuous-Integer Flow For Mixed-Integer Linear Programming

Fast Frank–Wolfe Algorithms with Adaptive Bregman Step-Size for Weakly Convex Functions

Video-KTR: Reinforcing Video Reasoning via Key Token Attribution

Non-Asymptotic Analysis of Efficiency in Conformalized Regression

Online Decision-Focused Learning

OVID: Open-Vocabulary Intrusion Detection

Neural Hamilton--Jacobi Characteristic Flows for Optimal Transport

From Fields to Random Trees

On Coreset for LASSO Regression Problem with Sensitivity Sampling

Randomization Boosts KV Caching, Learning Balances Query Load: A Joint Perspective

Weight Decay may matter more than µP for Learning Rate Transfer in Practice

Special Unitary Parameterized Estimators of Rotation

Reasoning Models Can be Accurately Pruned Via Chain-of-Thought Reconstruction

Making, Not Taking, the Best of N

VideoZoomer: Reinforcement-Learned Temporal Focusing for Long Video Reasoning

RECON: Robust symmetry discovery via Explicit Canonical Orientation Normalization

LoRA-Mixer: Coordinate Modular LoRA Experts Through Serial Attention Routing

Exposing Mixture and Annotating Confusion for Active Universal Test-Time Adaptation

CFT-RAG: An Entity Tree Based Retrieval Augmented Generation Algorithm With Cuckoo Filter

Mini-cluster Guided Long-tailed Deep Clustering

FlashAttention on a Napkin: A Diagrammatic Approach to Deep Learning IO-Awareness

2991. From Pixels to Semantics: Unified Facial Action Representation Learning for Micro-Expression Analysis

  • Topics: Other / Unclassified

Multi-Head Low-Rank Attention

Bayesian Attention Mechanism: A Probabilistic Framework for Positional Encoding and Context Length Extrapolation

Frayed RoPE and Long Inputs: A Geometric Perspective

Hilbert-Guided Sparse Local Attention

Measuring the Intrinsic Dimension of Earth Representations

Frequency Bands in RoPE: Base Frequency and Context Length Shape the Interpolation–Extrapolation Trade-off

VoMP: Predicting Volumetric Mechanical Property Fields

CARE: Covariance-Aware and Rank-Enhanced Decomposition for Enabling Multi-Head Latent Attention

MesaNet: Sequence Modeling by Locally Optimal Test-Time Training

Flatter Tokens are More Valuable for Speculative Draft Model Training

Continuous Chain of Thought Enables Parallel Exploration and Reasoning

Fine-Grained Activation Steering: Steering Less, Achieving More

A Law of Data Reconstruction for Random Features (And Beyond)

Avey-B

Back to Square Roots: An Optimal Bound on the Matrix Factorization Error for Multi-Epoch Differentially Private SGD

BézierFlow: Learning Bézier Stochastic Interpolant Schedulers for Few-Step Generation

Fingerprinting Deep Neural Networks for Ownership Protection: An Analytical Approach

Memorization Through the Lens of Sample Gradients

Interaction Field Matching: Overcoming Limitations of Electrostatic Models

MergePRAG: Orthogonal Merging of Passage-experts for Multi-hop Parametric RAG

Beyond Raw Detection Scores: Markov-Informed Calibration for Boosting Machine-Generated Text Detection

Why Do Unlearnable Examples Work: A Novel Perspective of Mutual Information

Sharpness-Aware Machine Unlearning

Erase or Hide? Suppressing Spurious Unlearning Neurons for Robust Unlearning

Not All Models Suit Expert Offloading: On Local Routing Consistency of Mixture-of-Expert Models

ContextBench: Modifying Contexts for Targeted Latent Activation and Behaviour Elicitation

In-Context Algebra

Reasoning or Retrieval? A Study of Answer Attribution on Large Reasoning Models

Testing Most Influential Sets

Directed Semi-Simplicial Learning with Applications to Brain Activity Decoding

Data-Aware and Scalable Sensitivity Analysis for Decision Tree Ensembles

ARMOR: Aligning Secure and Safe Large Language Models via Meticulous Reasoning

Formalising Human-in-the-Loop: Computational Reductions, Failure Modes, and Legal-Moral Responsibility

Breaking Gradient Temporal Collinearity for Robust Spiking Neural Networks

Mitigating Spurious Correlation via Distributionally Robust Learning with Hierarchical Ambiguity Sets

Diversity-Enhanced Reasoning for Subjective Questions

DIVERSE: Disagreement-Inducing Vector Evolution for Rashomon Set Exploration

When and Where to Reset Matters for Long-Term Test-Time Adaptation

Understanding the Role of Training Data in Test-Time Scaling

The Softmax Bottleneck Does Not Limit the Probabilities of the Most Likely Tokens

Computing Equilibrium beyond Unilateral Deviation

Decoupling Dynamical Richness from Representation Learning: Towards Practical Measurement

Testing Fourier Sparsity via Implicit Sensing

Stable coresets: Unleashing the power of uniform sampling

Random Label Prediction Heads for Studying Memorization in Deep Neural Networks

FedDAG: Clustered Federated Learning via Global Data and Gradient Integration for Heterogeneous Environments

Dynamical properties of dense associative memory

Learning in Prophet Inequalities with Noisy Observations

Personalized Collaborative Learning with Affinity-Based Variance Reduction

Preventing Model Collapse Under Overparametrization: Optimal Mixing Ratios for Interpolation Learning and Ridge Regression

Distilling to Hybrid Attention Models via KL-Guided Layer Selection

UniOD: A Universal Model for Outlier Detection across Diverse Domains

Light Differentiable Logic Gate Networks

Deep Learning with Learnable Product-Structured Activations

Video Unlearning via Low-Rank Refusal Vector

Deploying Models to Non-participating Clients in Federated Learning without Fine-tuning: A Hypernetwork-based Approach

Beyond Linear Processing: Dendritic Bilinear Integration in Spiking Neural Networks

Leveraging Discrete Function Decomposability for Scientific Design

MoMa: A Simple Modular Learning Framework for Material Property Prediction

A Scalable Constant-Factor Approximation Algorithm for WpW_p Optimal Transport

``Noisier'’ Noise Contrastive Estimation is (Almost) Maximum Likelihood

Curse of Slicing: Why Sliced Mutual Information is a Deceptive Measure of Statistical Dependence

Learning Collective Variables from BioEmu with Time-Lagged Generation

A Unifying View of Coverage in Linear Off-policy Evaluation

Robust Adaptive Multi-Step Predictive Shielding

Grouping Nodes with known Value Differences: A lossless UCT-based Abstraction Algorithm

Improving Extreme Wind Prediction with Frequency-Informed Learning

Welfarist Formulations for Diverse Similarity Search

Random Spiking Neural Networks are Stable and Spectrally Simple

Automatic and Structure-Aware Sparsification of Hybrid Neural ODEs with Application to Glucose Prediction

Decoupled Q-Chunking

ADM-v2: Pursuing Full-Horizon Roll-out in Dynamics Models for Offline Policy Learning and Evaluation

Riesz Neural Operator for Solving Partial Differential Equations

Beyond Structure: Invariant Crystal Property Prediction with Pseudo-Particle Ray Diffraction

Adaptive Mamba Neural Operators

Intention-Conditioned Flow Occupancy Models

Reliability-Adjusted Prioritized Experience Replay

Time Optimal Execution of Action Chunk Policies Beyond Demonstration Speed

ARTDECO: Toward High-Fidelity On-the-Fly Reconstruction with Hierarchical Gaussian Structure and Feed-Forward Guidance

AMPED: Adaptive Multi-objective Projection for balancing Exploration and skill Diversification

UrbanVerse: Scaling Urban Simulation by Watching City-Tour Videos

DemoGrasp: Universal Dexterous Grasping from a Single Demonstration

DexNDM: Closing the Reality Gap for Dexterous In-Hand Rotation via Joint-Wise Neural Dynamics Model

Language in the Flow of Time: Time-Series-Paired Texts Weaved into a Unified Temporal Narrative

Look-ahead Reasoning with a Learned Model in Imperfect Information Games

T1: One-to-One Channel-Head Binding for Multivariate Time-Series Imputation

Variance-Dependent Regret Lower Bounds for Contextual Bandits

Seq vs Seq: An Open Suite of Paired Encoders and Decoders

MAPSS: Manifold-based Assessment of Perceptual Source Separation

Information-based Value Iteration Networks for Decision Making Under Uncertainty

ReCAPA: Hierarchical Predictive Correction to Mitigate Cascading Failures

Calibrating Verbalized Confidence with Self-Generated Distractors

Group-Relative REINFORCE Is Secretly an Off-Policy Algorithm: Demystifying Some Myths About GRPO and Its Friends

Bounds of Chain-of-Thought Robustness: Reasoning Steps, Embed Norms, and Beyond

DeepScientist: Advancing Frontier-Pushing Scientific Findings Progressively

Neural Optimal Transport Meets Multivariate Conformal Prediction

Learning to Reason via Mixture-of-Thought for Logical Reasoning

DMAP: A Distribution Map for Text

A Balanced Neuro-Symbolic Approach for Commonsense Abductive Logic

Beyond Hearing: Learning Task-Agnostic ExG Representations from Earphones via Physiology-Informed Tokenization

DeepTRACE: Auditing Deep Research AI Systems for Tracking Reliability Across Citations and Evidence

Into the Rabbit Hull: From Task-Relevant Concepts in DINO to Minkowski Geometry

Not All Documents Are What You Need for Extracting Instruction Tuning Data

Revisiting Tree-Sliced Wasserstein Distance Through the Lens of the Fermat–Weber Problem

Mode-conditioning unlocks superior test-time compute scaling

Mixed-Curvature Tree-Sliced Wasserstein Distance

Text2Interact: High-Fidelity and Diverse Text-to-Two-Person Interaction Generation

Pursuing Minimal Sufficiency in Spatial Reasoning

ReIn: Conversational Error Recovery with Reasoning Inception

Towards a Theoretical Understanding of In-context Learning: Stability and Non-I.I.D Generalisation

Diverse Text Decoding via Iterative Reweighting

IGU-LoRA: Adaptive Rank Allocation via Integrated Gradients and Uncertainty-Aware Scoring

Learning from Label Proportions via Proportional Value Classification

Assembling the Mind's Mosaic: Towards EEG Semantic Intent Decoding

Advancing Spatiotemporal Representations in Spiking Neural Networks via Parametric Invertible Transformation

AlphaSAGE: Structure-Aware Alpha Mining via GFlowNets for Robust Exploration

Perturbation-Induced Linearization: Constructing Unlearnable Data with Solely Linear Classifiers

Robust Equation Structure Learning with Adaptive Refinement

MetaMuse: Algorithm Generation via Creative Ideation

Pushing on Multilingual Reasoning Models with Language-Mixed Chain-of-Thought

Learning to Play Multi-Follower Bayesian Stackelberg Games

SIM-CoT: Supervised Implicit Chain-of-Thought

AutoFigure: Generating and Refining Publication-Ready Scientific Illustrations

Measuring Uncertainty Calibration

EvolProver: Advancing Automated theorem proving by Evolving Formalized Problems via Symmetry and Difficulty

Distributionally Robust Linear Regression with Block Lewis Weights

Multi-LCB: Extending LiveCodeBench to Multiple Programming Languages

Improving Code Localization with Repository Memory

Bridging Piano Transcription and Rendering via Disentangled Score Content and Style

FedMC: Federated Manifold Calibration

Partial Soft-Matching Distance For Neural Representational Comparison With Partial Unit Correspondence

Neyman-Pearson Classification under Both Null and Alternative Distributions Shift

EasyCreator: Empowering 4D Creation through Video Inpainting

Geometry-aware Policy Imitation

Expressiveness of Multi-Neuron Convex Relaxations in Neural Network Certification

Sparse Attention Adaptation for Long Reasoning

QKV Projections Require a Fraction of Their Memory

Unlocking the Power of Co-Occurrence in CLIP: A DualPrompt-Driven Method for Training-Free Zero-Shot Multi-Label Classification

Multi-Synaptic Cooperation: A Bio-Inspired Framework for Robust and Scalable Continual Learning

Two-Layer Convolutional Autoencoders Trained on Normal Data Provably Detect Unseen Anomalies

Flow Map Learning Via Non-Gradient Vector Flow

A Relative Error-Based Evaluation Framework of Heterogeneous Treatment Effect Estimators

Learning Exposure Mapping Functions for Inferring Heterogeneous Peer Effects

Causal Discovery in the Wild: A Voting-Theoretic Ensemble Approach

Independence Test for Linear Non-Gaussian Data and Applications in Causal Discovery

Permutation-Consistent Variational Encoding for Incomplete Multi-View Multi-Label Classification

Self-Supervised Learning from Structural Invariance

Frequency-Domain Better than Time-Domain for Causal Structure Recovery in Dynamical Systems on Networks

Preserving Forgery Artifacts: AI-Generated Video Detection at Native Scale

Disentangled representation learning through unsupervised symmetry group discovery

There Was Never a Bottleneck in Concept Bottleneck Models

SNAPHARD CONTRAST LEARNING

UniEdit-Flow: Unleashing Inversion and Editing in the Era of Flow Models

Command-V: Training-Free Representation Finetuning Transfer

Merge before Forget: A Single LoRA Continual Learning via Continual Merging

∂∞\boldsymbol{\partial^\infty}-Grid: A Neural Differential Equation Solver with Differentiable Feature Grids

LayerSync: Self-aligning Intermediate Layers

Dataless Weight Disentanglement in Task Arithmetic via Kronecker-Factored Approximate Curvature

Geometric Autoencoder Priors for Bayesian Inversion: Learn First Observe Later

DiffWind: Physics-Informed Differentiable Modeling of Wind-Driven Object Dynamics

Articulation in Motion: Prior-free Part Mobility Analysis for Articulated Objects By Dynamic-Static Disentanglement

Uncertainty-Aware Diagnostics for Physics-Informed Machine Learning

VeriCoT: Neuro-symbolic Chain-of-Thought Validation via Logical Consistency Checks

CAD-Tokenizer: Towards Text-Based CAD Prototyping via Modality-Specific Tokenization

Resurfacing the Instance-only Dependent Label Noise Model through Loss Correction

Multi-Condition Conformal Selection

WinT3R: Window-Based Streaming Reconstruction with Camera Token Pool

Conformalized Decision Risk Assessment

HUMOF: Human Motion Forecasting in Interactive Social Scenes

Revisiting Active Sequential Prediction-Powered Mean Estimation

Angle K-Means

Accessible, Realistic, and Fair Evaluation of Positive-Unlabeled Learning Algorithms

Trace Anything: Representing Any Video in 4D via Trajectory Fields

Boosting for Predictive Sufficiency

Adaptive Conformal Guidance for Learning under Uncertainty

Bandits with Single-Peaked Preferences and Limited Resources

back arrowGo to TMLR homepage Slicing the Gaussian Mixture Wasserstein Distance

3802. UltraGauss: Ultrafast Gaussian Reconstruction of 3D Ultrasound Volumes

  • Topics: Computer Vision, Audio & Speech

Towards Better Branching Policies: Leveraging the Sequential Nature of Branch-and-Bound Tree

Learning to Adapt: In-Context Learning Beyond Stationarity

Improving Feasibility via Fast Autoencoder-Based Projections

Strongly Convex Sets in Riemannian Manifolds

Cambrian-S: Towards Spatial Supersensing in Video

SAM 3: Segment Anything with Concepts

ThinkOmni: Lifting Textual Reasoning to Omni-modal Scenarios via Guidance Decoding

FrameThinker: Learning to Think with Long Videos via Multi-Turn Frame Spotlighting

ARFlow: Auto-regressive Optical Flow Estimation for Arbitrary-Length Videos via Progressive Next-Frame Forecasting

Cautious Weight Decay

Event-T2M: Event-level Conditioning for Complex Text-to-Motion Synthesis

Test-time Verification via Optimal Transport: Coverage, ROC, & Sub-optimality

Riemannian Federated Learning via Averaging Gradient Streams

Robust Training of Neural Networks at Arbitrary Precision and Sparsity

T-TAMER: Provably Taming Trade-offs in ML Serving

Redirection for Erasing Memory (REM): Towards a universal unlearning method for corrupted data

Splat Regression Models

PPLLaVA: Varied Video Sequence Understanding With Prompt Guidance

FS-KAN: Permutation Equivariant Kolmogorov-Arnold Networks via Function Sharing

C-Voting: Confidence-Based Test-Time Voting without Explicit Energy Functions

Talking Points: Describing and Localizing Pixels

Tell me Habibi, is it Real or Fake?

Poly-attention: a general scheme for higher-order self-attention

Samples Are Not Equal: A Sample Selection Approach for Deep Clustering

CLIP-FMoE: Scalable CLIP via Fused Mixture-of-Experts with Enforced Specialization

UltraMemV2: Memory Networks Scaling to 120B Parameters with Superior Long-Context Learning

Textual Equilibrium Propagation for Deep Compound AI Systems

In Context Semi-Supervised Learning

Local Linear Attention: An Optimal Interpolation of Linear and Softmax Attention For Test-Time Regression

RACE Attention: A Strictly Linear-Time Attention for Long-Sequence Training

ASMIL: Attention-Stabilized Multiple Instance Learning for Whole-Slide Imaging

StPR: Spatiotemporal Preservation and Routing for Exemplar-Free Video Class-Incremental Learning

PSP: Prompt-Guided Self-Training Sampling Policy for Active Prompt Learning

GaitSnippet: Gait Recognition Beyond Unordered Sets and Ordered Sequences

Distributional Consistency Loss: Beyond Pointwise Data Terms in Inverse Problems

Maximizing Asynchronicity in Event-based Neural Networks

SceneStreamer: Continuous Scenario Generation as Next Token Group Prediction

Seeing What’s Wrong: A Trajectory-Guided Approach to Caption Error Detection

Contact Wasserstein Geodesics for Non-Conservative Schrödinger Bridges

The Gaussian-Head OFL Family: One-Shot Federated Learning from Client Global Statistics

SpecBranch: Speculative Decoding via Hybrid Drafting and Rollback-Aware Branch Parallelism

Fair Conformal Classification via Learning Representation-Based Groups

Open Data Synthesis for Deep Research

TrainRef: Curating Data with Label Distribution and Minimal Reference for Accurate Prediction and Reliable Confidence

AP-OOD: Attention Pooling for Out-of- Distribution Detection

Learning to Weight Parameters for Training Data Attribution

Evaluating Data Influence in Meta Learning

Robustify Spiking Neural Networks via Dominant Singular Deflation under Heterogeneous Training Vulnerability

Inverse Scaling in Test-Time Compute

4075. Bi-directional Bias Attribution: Debiasing Large Language Models without Modifying Prompts

  • Topics: LLMs & Foundation Models, Trust & Safety

FaLW: A Forgetting-aware Loss Reweighting for Long-tailed Unlearning

Parameterized Hardness of Zonotope Containment and Neural Network Verification

What Scales in Cross-Entropy Scaling Law?

Spectrum Tuning: Post-Training for Distributional Coverage and In-Context Steerability

Learning on a Razor’s Edge: Identifiability and Singularity of Polynomial Neural Networks

Residual Feature Integration is Sufficient to Prevent Negative Transfer

On the Bayes Inconsistency of Disagreement Discrepancy Surrogates

TangleScore: Tangle-Guided Purge and Imprint for Unstructured Knowledge Editing

BoRA: Towards More Expressive Low-Rank Adaptation with Block Diversity

General search techniques without common knowledge for imperfect-information games, and application to superhuman Fog of War chess

Understanding the Learning Phases in Self-Supervised Learning via Critical Periods

Bi-Criteria Metric Distortion

Learning-Augmented Moment Estimation on Time-Decay Models

Beyond Uniformity: Regularizing Implicit Neural Representations through a Lipschitz Lens

Quantum machine learning advantages beyond hardness of evaluation

Achieving Approximate Symmetry Is Exponentially Easier than Exact Symmetry

Toward Practical Equilibrium Propagation: Brain-inspired Recurrent Neural Network with Feedback Regulation and Residual Connections

Subquadratic Algorithms and Hardness for Attention with Any Temperature

Choices Speak Louder than Questions

Gistify: Codebase-Level Understanding via Runtime Execution

Soft Tokens, Hard Truths

The Price of Robustness: Stable Classifiers Need Overparameterization

Signal in the Noise: Polysemantic Interference Transfers and Predicts Cross-Model Influence

ReFeR: Improving Evaluation and Reasoning through Hierarchy of Models

4143. GeoGramBench: Benchmarking the Geometric Program Reasoning in Modern LLMs

  • Topics: LLMs & Foundation Models, Data-centric & Curation

Mapping Semantic & Syntactic Relationships with Geometric Rotation

Bilinear representation mitigates reversal curse and enables consistent model editing

Scaling Laws and Symmetry, Evidence from Neural Force Fields

Verifier-Constrained Flow Expansion for Discovery Beyond the Data

Causal Interpretation of Neural Network Computations with Contribution Decomposition

CDBridge: A Cross-omics Post-training Bridge Strategy for Context-aware Biological Modeling

Softmax is not Enough (for Adaptive Conformal Classification)

WFR-FM: Simulation-Free Dynamic Unbalanced Optimal Transport

HOTA: Hamiltonian framework for Optimal Transport Advection

Conjuring Semantic Similarity

Hierarchical Multi-Stage Recovery Framework for Kronecker Compressed Sensing

HistoPrism: Unlocking Functional Pathway Analysis from Pan-Cancer Histology via Gene Expression Prediction

Identity-Free Deferral For Unseen Experts

DRIFT-Net: A Spectral-Coupled Neural Operator for PDEs Learning

Reinforcement Mid-Training

CFO: Learning Continuous-Time PDE Dynamics via Flow-Matched Neural Operators

KANO: Kolmogorov-Arnold Neural Operator

Buckingham π\pi-Invariant Test‑Time Projection for Robust PDE Surrogate Modeling

Variational Pseudo Marginal Methods for Jet Reconstruction in Particle Physics

4261. TD-JEPA: Latent-predictive Representations for Zero-Shot Reinforcement Learning

  • Topics: Reinforcement Learning

Universal Value-Function Uncertainties

Is Pure Exploitation Sufficient in Exogenous MDPs with Linear Function Approximation?

Revisiting Matrix Sketching in Linear Bandits: Achieving Sublinear Regret via Dyadic Block Sketching

ARROW: An Adaptive Rollout and Routing Method for Global Weather Forecasting

GTM: A General Time-series Model for Enhanced Representation Learning of Time-Series data

Causal Score Conditioning for Multi-Resolution Latent Systems

Zero-Overhead Introspection for Adaptive Test-Time Compute

When to Retrain after Drift: A Data-Only Test of Post-Drift Data Size Sufficiency

Death of the Novel(ty): Beyond N-Gram Novelty as a Metric for Textual Creativity

Bayesian Ensemble for Sequential Decision-Making

SRFT: A Single-Stage Method with Supervised and Reinforcement Fine-Tuning for Reasoning

Breaking Barriers: Do Reinforcement Post Training Gains Transfer To Unseen Domains?

CONCUR: A Framework for Continual Constrained and Unconstrained Routing

SUSD: Structured Unsupervised Skill Discovery through State Factorization

Adaptive Conformal Prediction via Mixture-of-Experts Gating Similarity

AssoMem: Scalable Memory QA with Multi-Signal Associative Retrieval

ProfBench: Multi-Domain Rubrics requiring Professional Knowledge to Answer and Judge

Terminal Velocity Matching

Robust Decision-Making with Partially Calibrated Forecasters

Slicing Wasserstein over Wasserstein via Functional Optimal Transport

SCI-Verifier: Scientific Verifier with Thinking

C-Evolve: Consensus-based Evolution for Prompt Groups

POEMetric: The Last Stanza of Humanity

Statistical Guarantees for Offline Domain Randomization

Hallucination Reduction with CASAL: Contrastive Activation Steering for Amortized Learning

Learning From the Past with Cascading Eligibility Traces

Inferring brain plasticity rule under long-term stimulation with structured recurrent dynamics

Estimating Dimensionality of Neural Representations from Finite Samples

Setting up for failure: automatic discovery of the neural mechanisms of cognitive errors

Musculoskeletal simulation of limb movement biomechanics in Drosophila melanogaster

Video-STAR: Reinforcing Open-Vocabulary Action Recognition with Tools

Semi-Parametric Contextual Pricing with General Smoothness

ASSESS: A Semantic and Structural Evaluation Framework for Statement Similarity

How Many Code and Test Cases Are Enough? Evaluating Test Cases Generation from a Binary-Matrix Perspective

RATE-DISTORTION OPTIMIZED PRAGMATIC COMMUNICATION FOR COLLABORATIVE PERCEPTION

TEMPFLOW-GRPO: WHEN TIMING MATTERS FOR GRPO IN FLOW MODELS

RAG4DMC: Retrieval-Augmented Generation for Data-Level Modality Completion

Fathom-DeepResearch: Unlocking Long Horizon Information Retrieval and Synthesis for SLMs

Lean Finder: Semantic Search for Mathlib That Understands User Intents

On The Geometry and Topology of Representations: the Manifolds of Modular Addition

Quadratic Direct Forecast for Training Multi-Step Time-Series Forecast Models

Divide and Abstract: Autoformalization via Decomposition and Abstraction Learning

MuonBP: Faster Muon via Block-Periodic Orthogonalization

Poisson Midpoint Method for Log Concave Sampling: Beyond the Strong Error Lower Bounds

A Tale of Two Geometries: Adaptive Optimizers and Non-Euclidean Descent

Skirting Additive Error Barriers for Private Turnstile Streams

PERSISTENCE SPHERES: BI-CONTINUOUS REPRESENTATIONS OF PERSISTENCE DIAGRAMS.

Seesaw: Accelerating Training by Balancing Batch Size and Learning Rate Scheduling

PolySkill: Learning Generalizable Skills Through Polymorphic Abstraction For Continual Learning

Tree-sliced Sobolev IPM

Hierarchical Encoding Tree with Modality Mixup for Cross-modal Hashing

FinSearchComp: Towards a Realistic, Expert-Level Evaluation of Financial Search and Reasoning

Joint Distribution–Informed Shapley Values for Sparse Counterfactual Explanations

Conformalized Survival Counterfactuals Prediction for General Right-Censored Data

Joint Shadow Generation and Relighting via Light-Geometry Interaction Maps

Modeling Interference for Treatment Effect Estimation in Network Dynamic Environment

Identifiability Challenges in Sparse Linear Ordinary Differential Equations

Scalable Training for Vector-Quantized Networks with 100% Codebook Utilization

Causal Discovery via Quantile Partial Effect

Jacobian Aligned Random Forests

KLAS: Using Similarity to Stitch Neural Networks for Improved Accuracy-Efficiency Tradeoffs

SeeDNorm: Self-Rescaled Dynamic Normalization

Improving Set Function Approximation with Quasi-Arithmetic Neural Networks

MoRA: Mobility as the Backbone for Geospatial Representation Learning at Scale

OrthoRF: Exploring Orthogonality in Object-Centric Representations

Let OOD Feature Exploring Vast Predefined Classifiers

One-Shot Exemplars for Class Grounding in Self-Supervised Learning

ShieldedCode: Learning Robust Representations for Virtual Machine Protected Code

Optimizer Choice Matters For The Emergence of Neural Collapse

NeMo-map: Neural Implicit Flow Fields for Spatio-Temporal Motion Mapping

MergOPT: A Merge-Aware Optimizer for Robust Model Merging

FlyPrompt: Brain-Inspired Random-Expanded Routing with Temporal-Ensemble Experts for General Continual Learning

SimULi: Real-Time LiDAR and Camera Simulation with Unscented Transforms

How to Square Tensor Networks and Circuits Without Squaring Them

TESSAR: Geometry-Aware Active Regression via Dynamic Voronoi Tessellation

Light of Normals: Unified Feature Representation for Universal Photometric Stereo

Epistemic Uncertainty Quantification To Improve Decisions From Black-Box Models

A Federated Generalized Expectation-Maximization Algorithm for Mixture Models with an Unknown Number of Components

Dynamic Novel View Synthesis in High Dynamic Range

S2GO: Streaming Sparse Gaussian Occupancy

Neural Collapse in Multi-Task Learning

From Predictors to Samplers via the Training Trajectory

FlexLoRA: Entropy-Guided Flexible Low-Rank Adaptation

MT-DAO: Multi-Timescale Distributed Adaptive Optimizers with Local Updates

Understanding and improving Shampoo and SOAP via Kullback-Leibler Minimization

Beyond Sequential Reranking: Reranker-Guided Search Improves Reasoning Intensive Retrieval

FutureFill: Fast Generation from Convolutional Sequence Models

Displacement-Resistant Extensions of DPO with Nonconvex ff-Divergences

MRAD: Zero-Shot Anomaly Detection with Memory-Driven Retrieval

R-Horizon: How Far Can Your Large Reasoning Model Really Go in Breadth and Depth?

MARS - A Foundational Map Auto-Regressor

BEP: A Binary Error Propagation Algorithm for Binary Neural Networks Training

Hilbert: Recursively Building Formal Proofs with Informal Reasoning

Counterfactual Reasoning for Retrieval-Augmented Generation

LoFT: Low-Rank Adaptation That Behaves Like Full Fine-Tuning

Point2RBox-v3: Self-Bootstrapping from Point Annotations via Integrated Pseudo-Label Refinement and Utilization

Fostering Video Reasoning via Next-Event Prediction

ProxyAttn: Guided Sparse Attention via Representative Heads

RESA: Bringing Back What Sparse Attention Ignores with Residual Estimation

Training-Free Loosely Speculative Decoding: Accepting Semantically Correct Drafts Beyond Exact Match

PrefixMemory-Tuning: Modernizing Prefix-Tuning by Decoupling the Prefix from Attention

Interaction-aware Representation Modeling With Co-Occurrence Consistency for Egocentric Hand-Object Parsing

Sequential Parallel Duality in Prefix Scannable Models

Pareto Variational Autoencoder

CortiLife: A Unified Framework for Cortical Representation Learning across the Lifespan

Hot PATE: Private Aggregation of Distributions for Diverse Tasks

Rethinking Benign Relearning: Syntax as the Hidden Driver of Unlearning Failures

LoRA meets Riemannion: Muon Optimizer for Parametrization-independent Low-Rank Adapters

WARP: Weight Teleportation for Attack-Resilient Unlearning Protocols

Negative Pre-activations Differentiate Syntax

Heterogeneous Federated Fine-Tuning with Parallel One-Rank Adaptation

TriC-Motion: Tri-Domain Causal Modeling Grounded Text-to-Motion Generation

LUMINA: Detecting Hallucinations in RAG System with Context–Knowledge Signals

CUPID: A Plug-in Framework for Joint Aleatoric and Epistemic Uncertainty Estimation with a Single Model

DynaGuard: A Dynamic Guardian Model With User-Defined Policies

AtC: Aggregate-then-Calibrate for Human-centered Assessment

ProReGen: Progressive Residual Generation under Attribute Correlations

Any-Subgroup Equivariant Networks via Symmetry Breaking

A Bayesian Nonparametric Framework For Learning Disentangled Representations

Practical estimation of the optimal classification error with soft labels and calibration

The Value of Information in Human-AI Decision-making

SEED-SET: Scalable Evolving Experimental Design for System-level Ethical Testing

Is In-Context Learning Learning?

Mechanism of Task-oriented Information Removal in In-context Learning

SeRI: Gradient-Free Sensitive Region Identification in Decision-Based Black-Box Attacks

Rethinking Pareto Frontier: On the Optimal Trade-offs in Fair Classification

f-INE: A Hypothesis Testing Framework for Estimating Influence under Training Randomness

Tackling the XAI Disagreement Problem with Adaptive Feature Grouping

Faithful Bi-Directional Model Steering via Distribution Matching and Distributed Interchange Interventions

Counterfactual Explanations on Robust Perceptual Geodesics

No Prior, No Leakage: Revisiting Reconstruction Attacks in Trained Neural Networks

The Seismic Wavefield Common Task Framework

Play to Generalize: Learning to Reason Through Game Play

RIG: Synergizing Reasoning and Imagination in End-to-End Generalist Policy

Uncertainty Estimation via Hyperspherical Confidence Mapping

Composable Sparse Subnetworks via Maximum-Entropy Principle

Activation Function Design Sustains Plasticity in Continual Learning

PAS: Estimating the target accuracy before domain adaptation

Designing Rules to Pick a Rule: Aggregation by Consistency

Branched Schrödinger Bridge Matching

Good Allocations from Bad Estimates

Distilling Causal Signals for One-Shot Directed Evolution of Antibodies

Why High-rank Neural Networks Generalize?: An Algebraic Framework with RKHSs

Towards a Sharp Analysis of Offline Policy Learning for ff-Divergence-Regularized Contextual Bandits

Self-Supervised Evolution Operator Learning for High-Dimensional Dynamical Systems

Learning Escorted Protocols For Multistate Free-Energy Estimation

On the Spectral Differences Between NTK and CNTK and Their Implications for Point Cloud Recognition

Expressive Power of Implicit Models: Rich Equilibria and Test-Time Scaling

SAVE: A Generalizable Framework for Multi-Condition Single-Cell Generation with Gene Block Attention

Sublinear Spectral Clustering Oracle with Little Memory

Count Bridges enable Modeling and Deconvolving Transcriptomic Data

Two (narrow) heads are better than (an arbitrarily wide) one

Influence Dynamics and Stagewise Data Attribution

SAQ: Stabilizer-Aware Quantum Error Correction Decoder

Lean4Physics: Comprehensive Reasoning Framework for College-level Physics in Lean4

The Deleuzian Representation Hypothesis

Extending Fourier Neural Operators for Modeling Parameterized and Coupled PDEs

DADA: Dual Averaging with Distance Adaptation

Decision-Theoretic Approaches for Improved Learning-Augmented Algorithms

Beyond Short Steps in Frank-Wolfe Algorithms

Einstein Fields: A Neural Perspective To Computational General Relativity

Queue Length Regret Bounds for Contextual Queueing Bandits

Language Identification in the Limit with Computational Trace

The Expressive Limits of Diagonal SSMs for State-Tracking

Bird's-eye-view Informed Reasoning Driver

UnLoc: Leveraging Depth Uncertainties for Floorplan Localization

Masked Skill Token Training for Hierarchical Off-Dynamics Transfer

Towards Generalizable PDE Dynamics Forecasting via Physics-Guided Invariant Learning

Reference Grounded Skill Discovery

Zero-shot Forecasting by Simulation Alone

From Samples to Scenarios: A New Paradigm for Probabilistic Forecasting

PaAno: Patch-Based Representation Learning for Time-Series Anomaly Detection

Learning to Answer from Correct Demonstrations

Language and Experience: A Computational Model of Social Learning in Complex Tasks

ACADREASON: Exploring the Limits of Reasoning Models with Academic Research Problems

RuleReasoner: Reinforced Rule-based Reasoning via Domain-aware Dynamic Sampling

RefTool: Reference-Guided Tool Creation for Knowledge-Intensive Reasoning

What Happens Next? Anticipating Future Motion by Generating Point Trajectories

ProPerSim: Developing Proactive and Personalized AI Assistants through User-Assistant Simulation

Reinforcing General Reasoning Without Verifiers

WebWeaver: Structuring Web-Scale Evidence with Dynamic Outlines for Open-Ended Deep Research

IF-VidCap: Can Video Caption Models Follow Instructions?

Cannistraci-Hebb Training on Ultra-Sparse Spiking Neural Networks

Quasi-Equivariant Metanetworks

Data-to-Energy Stochastic Dynamics

Anchored Supervised Fine-Tuning

A General Framework for Black-Box Attacks Under Cost Asymmetry

Topology Matters in RTL Circuit Representation Learning

Faithfulness Under the Distribution: A New Look at Attribution Evaluation

LEGATO: Large-scale End-to-end Generalizable Approach to Typeset OMR

CollectiveKV: Decoupling and Sharing Collaborative Information in Sequential Recommendation

Selection, Reflection and Self-Refinement: Revisit Reasoning Tasks via a Causal Lens

Leveraging a Simulator for Learning Causal Representations from Post-Treatment Covariates for CATE

5336. Attack-Resistant Watermarking for AIGC Image Forensics via Diffusion-based Semantic Deflection

  • Topics: Diffusion Models & Generative AI, Computer Vision, Trust & Safety

The Natural Geometry of Code: Hyperbolic Representation Learning for Program Reasoning

A Two-Phase Deep Learning Framework for Adaptive Time-Stepping in High-Speed Flow Modeling

Discrete Adjoint Matching

Tversky Neural Networks: Psychologically Plausible Deep Learning with Differentiable Tversky Similarity

FACT: a first-principles alternative to the Neural Feature Ansatz for how networks learn representations

Adaptive Hopfield Network: Rethinking Similarities in Associative Memory

Referring Layer Decomposition