Skip to main content
PREreviews
-
Karmendra Pandey reviewed (Why) Is My Prompt Getting Worse? Rethinking Regression Testing for Evolving LLM APIs
- Software
- Information Systems
- Computer Networks and Communications
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Where Does Agent Reliability Come From? A Cross-Benchmark Decomposition of Verification Loops, Specialist Models, and Scaffolding in a Production Enterprise Agent
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Memory Control Signals Emerge Before Action in Long Horizon Agents
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Sandlock: Confining AI Agent Code with Unprivileged Linux Primitives
- Artificial Intelligence
- Sociology and Political Science
- Computer Networks and Communications
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Beyond Static Sandboxing: Learned Capability Governance for Autonomous AI Agents
- Artificial Intelligence
- Sociology and Political Science
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Orchestrating Intelligence: Confidence-Aware Routing for Efficient Multi-Agent Collaboration across Multi-Scale Models
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed REAP: Automatic Curation of Coding Agent Benchmarks from Interactive Production Usage
- Information Systems
- Software
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Where Does Agent Reliability Come From? A Cross-Benchmark Decomposition of Verification Loops, Specialist Models, and Scaffolding in a Production Enterprise Agent
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Beyond Static Sandboxing: Learned Capability Governance for Autonomous AI Agents
- Artificial Intelligence
- Sociology and Political Science
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Where Does Agent Reliability Come From? A Cross-Benchmark Decomposition of Verification Loops, Specialist Models, and Scaffolding in a Production Enterprise Agent
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed What AI Benchmarks Actually Measure: Adapting Convergent and Discriminant Validity to Interrogate Fifty-Six AI Benchmarks
- Safety Research
- Artificial Intelligence
- Management Science and Operations Research
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed What Should an Agent Remember? Disentangling Retention from Retrieval in Bounded-Memory Evaluation
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed VeriHarness: Scaling Agentic Verification for Long-Horizon Tasks
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Efficient Benchmarking in Production: A Study of an Evolving LLM Agent
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Towards a Science of AI Agent Reliability
- Artificial Intelligence
- Safety, Risk, Reliability and Quality
- Safety Research
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Towards a Science of AI Agent Reliability
- Artificial Intelligence
- Safety, Risk, Reliability and Quality
- Safety Research
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed TRACER: Trace-Based Adaptive Cost-Efficient Routing for LLM Classification
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Measured Joules, Learned Routes: Learning to Route for Energy-Efficient LLM Serving
- Electrical and Electronic Engineering
- Hardware and Architecture
- Artificial Intelligence
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed COMED: The Missing Middle Between Routing and Collaboration in Multi-LLM Inference
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Planner-as-Router: Joint Plan-Time Model Routing for Cost-Efficient Multi-Agent Workflows
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Pretrain Once, Route Anywhere: Towards a Foundation Model for LLM Routing
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed RSI-Router: Evolving Subtask-Level LLM Routing and Skills for Cost-Efficient Agents
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed The Complexity Trap: Simple Observation Masking Is as Efficient as LLM Summarization for Agent Context Management
- Information Systems
- Artificial Intelligence
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed FlexRouter: Learning Complementary Model Sets for Flexible LLM Routing
- Artificial Intelligence
- Information Systems
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed How Do AI Agents Spend Your Money? Analyzing and Predicting Token Consumption in Agentic Coding Tasks
- Information Systems
- Artificial Intelligence
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Routing Should Pay for Itself: Sparse Supervision for Economical LLM Routing
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed You Cannot Pick a Provider From the Price List: Market-Aware Routing for Open-Weight LLM Inference
- Computer Networks and Communications
- Electrical and Electronic Engineering
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Who judges the judges? Governance from metrics: a runtime framework for continuous LLM compliance monitoring
- Safety Research
- Artificial Intelligence
- Health Informatics
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed Argo-Bench: Evaluating Data Agents on Enterprise-Scale Workflows
- Management Information Systems
- Information Systems and Management
- Management Science and Operations Research
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed AutoCompact: Learning When to Compact Context in Long-Horizon Coding Agents
- Information Systems
- Software
- Artificial Intelligence
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed When Drift Detectors cry Wolf: False Alarm Rates in continuous ML Monitoring
- Review published
- Preprint server
- arXiv
-
Karmendra Pandey reviewed RepoRescue: An Empirical Study of LLM Agents on Whole-Repository Compatibility Rescue
- Information Systems
- Information Systems and Management
- Computer Science Applications
- Review published
- Preprint server
- arXiv