PaperHub 排行榜

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

LLMs Get Lost In Multi-Turn Conversation

Probabilistic Kernel Function for Fast Angle Testing

Text-to-3D by Stitching a Multi-view Reconstruction Network to a Video Generator

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

Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments

Generative Universal Verifier as Multimodal Meta-Reasoner

BIRD-INTERACT: Re-imagining Text-to-SQL Evaluation via Lens of Dynamic Interactions

Characterizing the Discrete Geometry of ReLU Networks

EditBench: Evaluating LLM Abilities to Perform Real-World Instructed Code Edits

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

Hubble: a Model Suite to Advance the Study of LLM Memorization

Is it Thinking or Cheating? Detecting Implicit Reward Hacking by Measuring Reasoning Effort

Learning with Dual-level Noisy Correspondence for Multi-modal Entity Alignment

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

Pre-training under infinite compute

Quotient-Space Diffusion Models

RealPDEBench: A Benchmark for Complex Physical Systems with Real-World Data

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

Speculative Actions: A Lossless Framework for Faster AI Agents

Spherical Watermark: Encryption-Free, Lossless Watermarking for Diffusion Models

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

Temporal superposition and feature geometry of RNNs under memory demands

The Art of Scaling Reinforcement Learning Compute for LLMs

Differentiable Model Predictive Control on the GPU

DiffusionNFT: Online Diffusion Reinforcement with Forward Process

In-Place Test-Time Training

In-the-Flow Agentic System Optimization for Effective Planning and Tool Use

InfoTok: Adaptive Discrete Video Tokenizer via Information-Theoretic Compression

Latent Particle World Models: Self-supervised Object-centric Stochastic Dynamics Modeling

Sequences of Logits Reveal the Low Rank Structure of Language Models

The Coverage Principle: How Pre-Training Enables Post-Training

How Do Transformers Learn to Associate Tokens: Gradient Leading Terms Bring Mechanistic Interpretability

Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?

Scaling In-the-Wild Training for Diffusion-based Illumination Harmonization and Editing by Imposing Consistent Light Transport

An analytic theory of creativity in convolutional diffusion models

Learning Linear Attention in Polynomial Time

Safety Alignment Should be Made More Than Just a Few Tokens Deep

Simplifying, Stabilizing and Scaling Continuous-time Consistency Models

Flowing Datasets with Wasserstein over Wasserstein Gradient Flows

Learning with Expected Signatures: Theory and Applications

Model Immunization from a Condition Number Perspective

Strategy Coopetition Explains the Emergence and Transience of In-Context Learning

Optimal Mistake Bounds for Transductive Online Learning

Spectral Perturbation Bounds for Low-Rank Approximation with Applications to Privacy

Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think

Artificial Kuramoto Oscillatory Neurons

BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions

Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models

SAM 2: Segment Anything in Images and Videos

Unlocking the Power of Function Vectors for Characterizing and Mitigating Catastrophic Forgetting in Continual Instruction Tuning

AutoGFM: Automated Graph Foundation Model with Adaptive Architecture Customization

EmbodiedBench: Comprehensive Benchmarking Multi-modal Large Language Models for Vision-Driven Embodied Agents

Learning Smooth and Expressive Interatomic Potentials for Physical Property Prediction

Multi-agent Architecture Search via Agentic Supernet

Cybench: A Framework for Evaluating Cybersecurity Capabilities and Risks of Language Models

Learning stochastic dynamics from snapshots through regularized unbalanced optimal transport

LoRA Done RITE: Robust Invariant Transformation Equilibration for LoRA Optimization

OLMoE: Open Mixture-of-Experts Language Models

Spread Preference Annotation: Direct Preference Judgment for Efficient LLM Alignment

Transformers Provably Solve Parity Efficiently with Chain of Thought

Cut Your Losses in Large-Vocabulary Language Models

SANA: Efficient High-Resolution Text-to-Image Synthesis with Linear Diffusion Transformers

TANGO: Co-Speech Gesture Video Reenactment with Hierarchical Audio Motion Embedding and Diffusion Interpolation

Turning Up the Heat: Min-p Sampling for Creative and Coherent LLM Outputs

Scaling and evaluating sparse autoencoders

ABKD: Pursuing a Proper Allocation of the Probability Mass in Knowledge Distillation via $\alpha$-$\beta$-Divergence

Beyond Self-Repellent Kernels: History-Driven Target Towards Efficient Nonlinear MCMC on General Graphs

Expected Variational Inequalities

Roll the dice & look before you leap: Going beyond the creative limits of next-token prediction

Suitability Filter: A Statistical Framework for Classifier Evaluation in Real-World Deployment Settings

The dark side of the forces: assessing non-conservative force models for atomistic machine learning

A multiscale analysis of mean-field transformers in the moderate interaction regime

Adjoint Schrödinger Bridge Sampler

Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free

ImageNet-trained CNNs are not biased towards texture: Revisiting feature reliance through controlled suppression

Improved Regret Bounds for Gaussian Process Upper Confidence Bound in Bayesian Optimization

Large Language Diffusion Models

Learning long range dependencies through time reversal symmetry breaking

Pan-LUT: Efficient Pan-sharpening via Learnable Look-Up Tables

QoQ-Med: Building Multimodal Clinical Foundation Models with Domain-Aware GRPO Training

Learning (Approximately) Equivariant Networks via Constrained Optimization

Algorithm Development in Neural Networks: Insights from the Streaming Parity Task

Can MLLMs Reason in Multimodality? EMMA: An Enhanced MultiModal ReAsoning Benchmark

Emergence in non-neural models: grokking modular arithmetic via average gradient outer product

Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs

Fundamental Bias in Inverting Random Sampling Matrices with Application to Sub-sampled Newton

High-Dimensional Prediction for Sequential Decision Making

How Do Large Language Monkeys Get Their Power (Laws)?

Network Sparsity Unlocks the Scaling Potential of Deep Reinforcement Learning

Neural Discovery in Mathematics: Do Machines Dream of Colored Planes?

On Differential Privacy for Adaptively Solving Search Problems via Sketching

Orthogonal Subspace Decomposition for Generalizable AI-Generated Image Detection

Polynomial-Delay MAG Listing with Novel Locally Complete Orientation Rules

Retrieval-Augmented Perception: High-resolution Image Perception Meets Visual RAG

rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinking

Scaling Collapse Reveals Universal Dynamics in Compute-Optimally Trained Neural Networks

STAIR: Improving Safety Alignment with Introspective Reasoning

SWE-Lancer: Can Frontier LLMs Earn $1 Million from Real-World Freelance Software Engineering?

VideoRoPE: What Makes for Good Video Rotary Position Embedding?