archive
Every paper Pith has read. Search by title, abstract, or pith.
48309 papers indexed · page 7
-
Training-free method stretches video generation to full minutes
Head Forcing: Long Autoregressive Video Generation via Head Heterogeneity
-
GAN upsampling plus expert fusion cuts artifact bias in image detectors
Reduce the Artifacts Bias for More Generalizable AI-Generated Image Detection
-
Proportional liability rules guide agents to efficient shock paths
Efficient liability assignment under shock propagation
-
14 discrete phases suffice for mode-pairing QKD
Discrete-phase-randomized mode-pairing quantum key distribution
-
Counterfactual edits train better multi-agent orchestrators
LEMON: Learning Executable Multi-Agent Orchestration via Counterfactual Reinforcement Learning
-
WHTDM outperforms OFDM by 10x in BER at 120 km/h
WHTDM: Walsh-Hadamard Transform Division Multiplexing for Doubly-Selective Channels
-
Subband learning cuts DBP complexity for wideband fiber links
ML-assisted Subband Learned Digital Backpropagation for Nonlinearity Compensation in Wideband Optical Systems
-
Ming glossaries used flexible Chinese characters to approximate foreign sounds
Cross-Linguistic Transcription and Phonological Representation in the Hu\`it\'onggu\v{a}nx\`i Hu\'ay\'iy\`iy\v{u}
-
Dust drives acceleration in f(R,T) model within Hubble window
Energy conditions in consistent perfect fluid cosmology
-
Stale code snippets make models output outdated helpers
When Retrieval Hurts Code Completion: A Diagnostic Study of Stale Repository Context
-
Models evolve reusable skills library at test time
Test-Time Learning with an Evolving Library
-
Proof nets canonically represent PiL sequent derivations
Proof Nets for PiL (Full Version)
-
GeoVista plans globally then inspects branches for satellite images
GeoVista: Visually Grounded Active Perception for Ultra-High-Resolution Remote Sensing Understanding
-
Four-channel scheme reduces laser noise in Rydberg sensors
Weight Hybrid Architecture of Rydberg-Atomic Sensors
-
Probe shows RAG follows wrong context in 85 percent of conflict cases
Does RAG Know When Retrieval Is Wrong? Diagnosing Context Compliance under Knowledge Conflict
-
Higher-order expressions update electron-phonon total energy for all materials
Accurate computation of the electron-phonon interaction contribution to the total energy
-
Database unifies 77k high-pressure structures under single DFT standard
High-Pressure Crystal Structure Database
-
Localization theorem holds for motivic homotopy on analytic stacks
The Localization Theorem for the Motivic Homotopy Theory of Complex Analytic Stacks and other Geometric Settings
-
Geodesic currents with strongly hyperbolic duals are dense
Geodesic currents of coarse negative curvature
-
Metasurface doublet cuts SLM pixel pitch threefold
Complex wavefront engineering via decoupled space-time modulation
-
Focused estimator improves PU learning on imbalanced data
Focused PU learning from imbalanced data
-
New meter measures Poynting vector in light beams
Determination of Poynting Vector Characteristics
-
Cell-mask diffusion planner lifts table reasoning accuracy 15.76 points
From Table to Cell: Attention for Better Reasoning with TABALIGN
-
Retrieval augments schema graphs for relational database predictions
From Schema to Signal: Retrieval-Augmented Modeling for Relational Data Analytics
-
Kernel aggregation detects network changes without knowing the pattern
KAP-CPD: Kernel Aggregation for Change-Point Detection in Dynamic Networks
-
Human motion anchors objects in simulation-ready HOI from video
Real2Sim in HOI: Toward Physically Plausible HOI Reconstruction from Monocular Videos
-
User clicks remove objects in diffusion models without masks
ClickRemoval: An Interactive Open-Source Tool for Object Removal in Diffusion Models
-
Disguised compliance rules let attackers hijack LLM agents
Exploiting LLM Agent Supply Chains via Payload-less Skills
-
ReLU networks double P1 FEM rates for singular perturbations
Neural Networks for Singular Perturbations -- Finite Regularity
-
OmniDrop cuts omni-modal LLM latency by up to 40%
OmniDrop: Layer-wise Token Pruning for Omni-modal LLMs via Query-Guidance
-
Replaying early insights lifts accuracy in every LLM reasoning test
Stateful Reasoning via Insight Replay
-
Dunkl-Schrödinger Hardy dual is BMO subspace
The dual of the Hardy space associated to the Dunkl-Schr\"odinger operator with reverse H\"older class potential
-
IIQ metric scores AI integration from 0 to 1000
Intelligence Impact Quotient (IIQ): A Framework for Measuring Organizational AI Impact
-
Guardrails adapt from sparse noisy failures via conservative induction
LiSA: Lifelong Safety Adaptation via Conservative Policy Induction
-
Shared-bound assumption enables sparse precision estimates for interval data
Estimating Precision Matrices for High-Dimensional Interval-Valued Data
-
Global solutions exist for power-rate fragmentation-coagulation in any dimension
Local and global solutions to continuous fragmentation-coagulation equations with vanishing diffusion and unbounded fragmentation and coagulation rates
-
CP-OFDM yields lower ranging CRB than frequency-spread waveforms for large N
CP-OFDM Achieves Lower Ranging CRB Than Frequency-Spread Waveforms in the Large-Sample Regime
-
Distilled rerankers match quality with 34% fewer tokens
Stop Overthinking: Unlocking Efficient Listwise Reranking with Minimal Reasoning
-
Orthogonal projection isolates hallucination signals in LLM answers
When Answers Stray from Questions: Hallucination Detection via Question-Answer Orthogonal Decomposition
-
Adaptive gate skips reasoning for simple multimodal inputs
Think When Needed: Adaptive Reasoning-Driven Multimodal Embeddings with a Dual-LoRA Architecture
-
Bayesian analysis pins light dark matter ranges in neutron stars
Bayesian analysis of density profile of light dark matter elucidating the properties of dark matter admixed neutron stars in the presence of hyperons
-
Hardy-Littlewood asymptotics hold for integer points in higher-dimensional simplices
Integer points in a simplex and related Diophantine problems: Hardy--Littlewood asymptotics in higher dimensions
-
Synthesized open-ended problems raise LLM coding scores by 8-12 points
FrontierSmith: Synthesizing Open-Ended Coding Problems at Scale
-
Row sums on Gram overlaps recover inliers at sqrt(n) scale
Inlier Recovery for Robust Registration via Gram-Matrix Overlap
-
RL prompter lifts frozen LLM reasoning from 55% to 90%
Prompting Policies for Multi-step Reasoning and Tool-Use in Black-box LLMs with Iterative Distillation of Experience
-
4B genome agent matches larger LLMs on microbial trait prediction
GGBound: A Genome-Grounded Agent for Microbial Life-Boundary Prediction
-
One beam carries three distinct Skyrmion numbers
Tunable spatio-spectral Target Skyrmions and topological multiplexing
-
Sampling and model-checking oracles learn regular POMDP policies
Synthesizing POMDP Policies: Sampling Meets Model-checking via Learning
-
Oscillations add stabilizing repulsion to tetraquarks
Kapitza Dynamics as a New Stabilization Mechanism for Heavy Tetraquarks
-
Binary masks cut MoE FLOPs by 85% with 98% accuracy kept
BEAM: Binary Expert Activation Masking for Dynamic Routing in MoE