archive
Every paper Pith has read. Search by title, abstract, or pith.
85880 papers indexed · page 138
-
Context graphs guide LLMs to resolve code merge conflicts better
Rover: Context-aware Conflict Resolution with LLM
-
Cycle consistency ensures unique answers in generated reasoning tasks
A2RBench: An Automatic Paradigm for Formally Verifiable Abstract Reasoning Benchmark Generation
-
Non-unique nuclear coordinates reveal clusters in light nuclei
Emergence of Cluster Formation in Light Nuclei
-
Self-supervised models scale ECG performance reliably with size
How Do Electrocardiogram Models Scale?
-
Hybrid model meets all ES tests for equity risk forecasts
A Hybrid Gaussian Process Regression Framework for Stable Volatility-Covariance Estimation: Evidence from Global Equity Indices
-
Construction yields infinitely many modular lattices with complementation
Varieties and quasivarieties of lattices with complementation
-
Logit scores estimate client contributions per class in federated learning
Data-Free Client Contribution Estimation via Logit Maximization for Federated Learning
-
Over half of top AI model wins fail basic superiority tests
Position: State-of-the-Art Claims Require State-of-the-Art Evidence
-
Closed-form BER picks optimal angle for light-trail links
ISI Modeling and BER Performance for Rotating Light-Trail Image Sensor Communication
-
KL relaxation scales bi-causal transport via policy gradients
Scalable Bi-causal Optimal Transport via KL Relaxation and Policy Gradients
-
SymTrack boosts scene text tracking accuracy by up to 12%
Beyond Detection: A Structure-Aware Framework for Scene Text Tracking
-
Bregman framework gives U-calibration for Tsallis losses
Calibeating for general proper losses: A Bregman divergence approach
-
VLA driving explanations match scenes less than half the time
Is VLA Reasoning Faithful? Probing Safety of Chain-of-Causation in Autonomous Driving Models
-
Bypass gaps after unlearning do not confirm hidden memorization
Auditing Reasoning-Trace Memorization Claims after Unlearning with Head-Conditioned Canaries
-
Interleaving reviews with items improves generative recs
RAGR: Review-Augmented Generative Recommendation
-
Deep RL clusters cell-free networks from single channel estimates
Leveraging Deep Reinforcement Learning for Clustered Cell-Free Networking Over User Mobility
-
CliffSplit uncovers 15% higher error in molecular property cliffs
When Molecular Similarity Works: Property Cliffs Reveal Hidden Errors
-
Stretch-ICP cuts velocity errors 95% in aggressive lidar scans
Stretch-ICP: A Continuous-Trajectory Registration and Deskewing Algorithm in Scenarios of Aggressive Motions
-
Expert dashboards model cognition for AI education
Expert Cognition Dashboard: From Learning Analytics to Cognition Intelligence in AI-Driven Education
-
Multimodal AI struggles to infer user internal states
EgoIntrospect: An Egocentric Dataset and Benchmark for User-Centric Internal State Reasoning
-
BLAST workflow plus dual filters boosts protein QA for new sequences
Unlocking Biological Workflows for Robust Protein-Text Question Answering: A Dual-Dimensional RAG Framework
-
Compact encoder processes 8x more video frames at lower latency
LiteFrame: Efficient Vision Encoders Unlock Frame Scaling in Video LLMs
-
AI artifacts turn transcripts into shared maps and scores for better group analysis
CLARA: An AI-Augmented Analytics Dashboard for Collaboration Literacy
-
Dwarf nova disk develops two-armed flaring pattern
A growth of early superhump: Multi color observation of the WZ Sge star TCP J23580961+5502508
-
Nil 3-manifolds wrap with flexible exponent 8/3
Flexible exponent of geometric 3-manifolds and Legendrian maps of Seifert spaces
-
DL models detect grid faults in 15 ms but lag at 50-90 ms end-to-end
Latency-Aware Deep Learning Benchmark for Real-Time Cyber-Physical Attack and Fault Classification in Inverter-Dominated Power Grids
-
Benchmark adds 1000 Lean proof targets from applied math
CAM-Bench: A Benchmark for Computational and Applied Mathematics in Lean
-
One model unifies catalyst energy prediction and structure design
CatalyticMLLM: A Graph-Text Multimodal Large Language Model for Catalytic Materials
-
One model predicts and designs catalytic materials in one pass
CatalyticMLLM: A Graph-Text Multimodal Large Language Model for Catalytic Materials
-
Four dwarf galaxies appear dark-matter deficient
HI Observations of Baryon-Dominated Dwarf Galaxy Candidates
-
Shading and motion model improves depth on 2D screens
Monocular Depth Perception Enhancement Based on Joint Shading/Contrast Model and Motion Parallax (JSM)
-
L1 sandwiching enables quasipolynomial PQ learning for DNFs
Iterative Chow Filtering for Learning with Distribution Shift
-
Frequency calibration matches TTA adapters using far fewer parameters
Towards Principled Test-Time Adaptation for Time Series Forecasting
-
Dual spatial system tops VLN benchmarks
SEDualVLN: A Spatially-Enhanced Dual-System for Vision-Language Navigation
-
Four designs shape diffusion image-to-video models
Image-to-Video Diffusion: From Foundations to Open Frontiers
-
TIDE uses trial and debate to stabilize prompt optimization
Towards Robust Argumentative Essay Understanding via TIDE: An Interactive Framework with Trial and Debate
-
Fidelity probes lift spec-code agreement from 0.63 to 0.94
Fidelity Probes for Specification--Code Alignment
-
Random Forest detects telecom fraud at 99.9% accuracy
An Efficient Machine Learning-based Framework for Detection and Prevention of Frauds in Telecom Networks
-
DFM unifies one-step drift generation with multi-step refinement
Drift Flow Matching
-
Waveform misalignment breaks gradient flow in oscillator Ising machines
Breakdown of Gradient-Flow Dynamics in Oscillator Ising Machines from Harmonic Misalignment
-
Harmonic misalignment breaks gradient flow in oscillator Ising machines
Breakdown of Gradient-Flow Dynamics in Oscillator Ising Machines from Harmonic Misalignment
-
Automated TDD lifts AI web app success by 34-48 points
From Runnable to Shippable: Multi-Agent Test-Driven Development for Generating Full-Stack Web Applications from Requirements
-
IR frequency shift saturates Raman rectification at high pump power
Frequency renormalization and its effects in nonlinear phononics with $Q_RQ_{IR}^{2}$-type coupling
-
FORSS gives accurate power formulas for win stats on hierarchical endpoints
The FORSS Framework for Sample Size and Power Calculations With Win Statistics for Hierarchical Endpoints
-
Uncertainty handling anchors advanced control theory in practice
Handling Control System Uncertainty
-
Regret-optimal algorithms for position-aware MNL bandits
Learning in Position-Aware Multinomial Logit Bandits: From Multiplicative to General Position Effects
-
Refined Schwarz estimates yield boundary lemmas in Banach spaces
Some sharp Schwarz type estimates and their applications in Banach spaces
-
Vision transformer hits 95 percent on cervical cell classification
Systematic Evaluation of Vision Transformers for Automated Cervical Cancer Classification: Optimization, Statistical Validation, and Clinical Interpretability
-
Singular value functions now defined on every C*-algebra
Singular value functions for C\(^*\)-algebras
-
Surrogate-guided halving cuts scaling-law cost by 98.7 percent
Active Budget Allocation for Efficient Scaling Law Estimation via Surrogate-Guided Pruning