84,648Total entries

1Added by human

84,647Found by agent

12Categories

Knowledge catalogue

All entries

59,851 results

Research

What do Reward Models Memorize?

DGX agent

arXiv:2607.24484v1 Announce Type: cross Abstract: This paper studies what discriminatively trained reward models (RMs) memorize by measuring counterfactual memorization on two human preference dataset

researcharxiv-cs-cl

28 Jul 2026

Research

What Softmax Throws Away: Mass-Aware Attention for Evidence Accumulation

DGX agent

arXiv:2607.22781v1 Announce Type: cross Abstract: High task performance does not show whether a model retains prediction-relevant structural information in its internal representation. Temporal graph

All entries

What do Reward Models Memorize?

What Softmax Throws Away: Mass-Aware Attention for Evidence Accumulation

When Activation Oracles Learn Not to Read: Concept-Specific Blind Spots in Fine-Tuned Oracles

When Can Depth Replace Precision? A Resource Theory of Quantized Neural Computation

When Can You Correct Distribution Drift in Temporal Graph Generation? A Sharpening--Drift Tension and an Impossibility for Observation-Based Correction

When Every Simulation Counts: Value-Based Reinforcement Learning for Accelerated Photonics Inverse Design

When Less Is More: A Controlled Benchmark of Lightweight CNNs for Satellite Land-Cover Segmentation on DeepGlobe

When LLM Defenses Backfire: Characterizing Safety, Performance, and Cost Trade-offs

When Low CER is Not Enough: An Analysis of Hallucinations in Vision-Language OCR Systems on Historical Uruguayan Documents

When Rates Are Geometric: Rate-Certificate Transfer for Contact Splittings in Optimization

When Should Active RAG Retrieve? A Budget-Aware Evaluation of Utility, Calibration, and Cost

Where Is the Cost of Third-Party API Routers in Agentic Software Development?

Where Quality Breaks in Compressed Short-Text Generation: Staged Bottleneck Localization

Which Models Perform Better in Inheritance Reasoning?

Which Workloads Belong in Orbit? A Workload-First Framework for Orbital Data Centers Using Semantic Abstraction

Who Gets Named: Citation Type Predicts Individual Naming by Grounded Language Models, and a Roster Instrument Captures 0.5% of It

Who Pays the Price? Stakeholder-Centric Prompt Injection Benchmarking for Real-world Web Agents

Why does Greedy Search produce Optimal Clustering Outcomes? A Fixed-Core Assignment Theory

Wired for Overconfidence: A Mechanistic Perspective on Inflated Verbalized Confidence in LLMs

WISERouter: LLM Routing with Workload Budget Constraint

WorldDiT: A Unified Diffusion Architecture for World and Action Modeling

Wrong Design Intent Is Worse Than None: A Derangement-Control Diagnosis of Header Conditioning in CAD Program Completion

X-Stage: An Overlooked Pipeline Stage for Communication-Computation Overlap in DiT Inference

XGRVFL-MV: Residual-Coupled Graph-Embedded Multi-View Random Vector Functional Link Network with FleXi Guardian Loss

XMatchAD: A Cross-Modal Matching Perspective on Reconstruction-based Anomaly Detection

XMix: Combating Extremely Noisy Labels via Local Smoothness in Self-Supervised Feature Space

xMIx: High-Performance Serving-Time Platform for Mechanistic Interpretability Apps

You Talkin to Me?: A Network Analysis of Gendered Speaker-Addressee Patterns in Film Screenplays

Zing: Social Mind for LLMs

A Comparative Benchmark of Federated Learning Strategies for Mortality Prediction on Heterogeneous and Imbalanced Clinical Data

A Consensus-Based Framework for Relative Preference Evaluation of Large Language Models

A Defense of the Quadratic Model

A Drift Stable Quantum Federated Learning for Intelligent Services

A Dual Path Framework with Hotspot Guided Fusion for Three Dimensional CT to PET Synthesis in Head and Neck Cancer

A Factorial Study of Synthetic Data Generation for Low-Resource Machine Translation using Grammar Books

A Framework for Individual Tree Growth Reconstruction Using Multi-Platform Laser Scanning

A Graph-Based Control Interface for Traffic Signals on Heterogeneous Road Networks

A Leakage-Free Stacked Ensemble Method for Multiclass Classification

A Monolithic Hand with Asymmetric Origami Bending and Dual-chamber Actuators

A Robust Pipeline for Differentially Private Federated Learning on Imbalanced Clinical Data using SMOTETomek and FedProx

A Smooth Phase-Separation Model for Weak-Boundary Segmentation of Homogeneous Structures

A Unified Framework for Automated Assembly Sequence and Production Line Planning using Graph-based Optimization

ACME: A Multi-Cultural, Multi-Embodiment Social-Navigation Dataset

Action-Conditioned World Model for Goal Plane Probe Guidance in Robotic Ultrasound

Active few-shot segmentation by reinforcing data selection

AdamNX: An Adam improvement algorithm based on a novel exponential decay mechanism for the second-order moment estimate

Adaptive Contrast Enhancement and Optimised Feature Matching for RootSIFT-Based Palm-Vein Recognition

Adaptive Learned State Estimation based on KalmanNet