archive
Every paper Pith has read. Search by title, abstract, or pith.
85880 papers indexed · page 130
-
Causal tests select better memories for long-running AI agents
Causal Intervention-Based Memory Selection for Long-Horizon LLM Agents
-
Three-stage pipeline lifts video RAG retrieval from 0.195 to 0.759 nDCG
MARQUIS: A Three-Stage Pipeline for Video Retrieval-Augmented Generation
-
Co-citation predictability drops over 20 years
Temporal Decay of Co-Citation Predictability: A 20-Year Statute Retrieval Benchmark from 396M Ukrainian Court Citations
-
Quantum trajectories reach higher decoding scores than classical ones
Quantum Viterbi Algorithm
-
System maps hand contacts to surfaces in operating rooms
TouchMap-OR: Multi-View 3D Mapping of Hand-Surface Contacts
-
Agents reach 77% usable browser games but only 20% excellent
WebGameBench: Requirement-to-Application Evaluation for Coding Agents via Browser-Native Games
-
Agents reach 77% playable games but only 20% excellent
WebGameBench: Requirement-to-Application Evaluation for Coding Agents via Browser-Native Games
-
Fermi detects 0.36-degree gamma-ray glow near Berkeley 87
Diffuse gamma-ray emission in the vicinity of open cluster Berkeley 87
-
cGAN simulates FARICH responses faster than Monte Carlo
ML-based Fast Simulation of FARICH Responses
-
Adversary can always reframe prompt injections as legitimate
AI Agents May Always Fall for Prompt Injections
-
SparseSAM cuts SAM inference time in half with tiny accuracy loss
SparseSAM: Structured Sparsification of Activations in Segment Anything Models
-
Flexural waves drive transverse currents at double frequency in graphene
Finite-frequency anomaly-induced electromechanical response of Dirac fermions in deformed graphene
-
Variable latent heat gives parabolic melting curves
Theory of melting lines with a variable enthalpy of fusion
-
Point prompts from DINO prototypes raise SAM3 mIoU by 34 points on new farm images
SegRAG: Training-Free Retrieval-Augmented Semantic Segmentation
-
Point prompts from reference prototypes lift segmentation by 34 mIoU on farm images
SegRAG: Training-Free Retrieval-Augmented Semantic Segmentation
-
DL tunes pinching and movable antennas to lift ISAC sum rates
DL-Driven Optimization for ISAC System Equipped With Pinching and Movable Antennas
-
QUBO without penalties cuts chain breaks to 0.04 percent on D-Wave
A Penalty-Free Pipeline for Direct Quantum-Annealer Portfolio Optimization
-
Alignment strength alone creates multiple active phases
Collective dynamics of active matter with orientation-weighted alignment
-
Guided checks at code boundaries boost translation pass rates
Verifier-Guided Code Translation via Meta-Step Decoding
-
Dual memory sustains AI accuracy past 10,000 messages
Episodic-Semantic Memory Architecture for Long-Horizon Scientific Agents
-
FixMatch boosts multi-task vision models on partial labels
Multi-task learning on partially labeled datasets via invariant/equivariant semi-supervised learning
-
Hybrid portfolio solver wins via classical pipeline
Where the Quantum Lives in D-Wave Hybrid Portfolio Optimization: An Operational Decomposition Audit
-
Prime gaps obey iterative map from cohomological structure
Iterative maps emerging from cohomological structure of primes
-
Boundary fluctuations yield state-dependent Langevin noise
Variational Boundary Fluctuations as a First-Principles Origin of Langevin Noise
-
Wearable sensors forecast autism behaviors 10 minutes early
Prediction of Challenging Behaviors Associated with Profound Autism in a Classroom Setting Using Wearable Sensors
-
Wearables forecast autism behaviors 10 minutes early in class
Prediction of Challenging Behaviors Associated with Profound Autism in a Classroom Setting Using Wearable Sensors
-
GraphMind turns resolution traces into self-updating action graphs
GraphMind: From Operational Traces to Self-Evolving Workflow Automation
-
STVG collapse forms RN-like horizons in finite proper time
Junction Conditions and Gravitational Collapse in Scalar-Tensor-Vector Gravity
-
Shear alignment raises rod dispersion in tubes by up to 30%
Shear alignment and tensorial Taylor--Aris dispersion of Brownian rods in a circular tube
-
Compressed KV cache yields full accuracy at 4x throughput
VeriCache: Turning Lossy KV Cache into Lossless LLM Inference
-
CFS and GA tuning lift fault prediction accuracy to 88.4%
A Feature-Driven Framework for Software Fault Prediction
-
Fast-slow video guardrail tops larger models at lower cost
SafeLens: Deliberate and Efficient Video Guardrails with Fast-and-Slow Screening
-
Adaptive sampling cuts LLM verification cost to constant factor of optimum
Adaptive Generate-Rank-Verify: Inference-Time Search with Costly Verification
-
Bayesian-Monte Carlo method refines construction schedule forecasts
Bayesian-Monte Carlo Schedule Updating for Construction Digital Twins: A Probabilistic Framework for Dynamic Project Forecasting
-
RRM algorithms converge to efficient Bayes-Nash equilibrium
Convergence of Stochastic First-Order Algorithms in Bertrand Competition Under Incomplete Information
-
Regularization keeps NTK constant for classification in wide networks
The Neural Tangent Kernel for Classification
-
VENOM toolkit unifies generative models under one PyTorch interface
Venom: A PyTorch Generative Modeling Toolkit
-
Reverse annealing brackets device dynamics between two classical samplers
Schedule-dependent basin occupation in a programmable quantum annealer
-
Cycled reverse annealing brackets classical simulation results
Schedule-dependent basin occupation in a programmable quantum annealer
-
Zero-shot model downscales precipitation to daily 10 km fields globally
Longwang: Zero-Shot Global Spatiotemporal Precipitation Downscaling with a Latent Generative Prior
-
AutoRubric-T2I builds reward models from under 0.01% of preference data
AutoRubric-T2I: Robust Rule-Based Reward Model for Text-to-Image Alignment
-
AutoRubric-T2I turns tiny preference data into a compact set of explicit rubrics that let…
AutoRubric-T2I: Robust Rule-Based Reward Model for Text-to-Image Alignment
-
One demo yields general policy for contact-rich robot tasks
From a Single Demonstration to a General Policy for Contact-Rich Manipulation
-
8x8 transfer matrix solves three-chain Ising tube with 20 couplings
Exact solution and pair correlation functions for a generalized three-chain Ising tube with multispin interactions
-
Directional tolerances let inexact adjoint gradients drive airfoil optimization
Inexact Adjoint Gradients and Directional Tolerances for Full-Potential Airfoil Optimization
-
MoE models show deep-layer routing collapse for low-resource languages
Mixture of Experts for Low-Resource LLMs
-
Gray-box data enables distributed H2 controller design
Distributed Synthesis of Gray-Box Distributed H2 Controllers
-
Hybrid memory system keeps LLM agents consistent across sessions
NeuSymMS: A Hybrid Neuro-Symbolic Memory System for Persistent, Self-Curating LLM Agents
-
Hybrid system keeps LLM agent memory accurate across sessions
NeuSymMS: A Hybrid Neuro-Symbolic Memory System for Persistent, Self-Curating LLM Agents
-
Davenport constant determines elasticity of orders with prime conductor
Elasticity of Orders from the $S$-relative Davenport Constant: an Arithmetic Application of a Number-Theoretic Investigation