Skip to main content
PREreviews
-
Evgenii Arsentev reviewed Compaction Laundering: Provenance Destruction in Context-Compacting Coding Agents
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed REM: What to Store, When to Recall, What to Forget in Experience Memory for Coding Agents
- Information Systems
- Artificial Intelligence
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed When Coding Agents Outpace Review: Evaluating PRD-Driven Development in Mature Enterprise Repositories
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed Skip the Compressor: Coding Agents Fix More Issues When They Pick Whole Files
- Information Systems
- Software
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed The Role of Agentic Instruction Categories for Code Generation: A Study of Five Python Libraries
- Information Systems
- Software
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed CodeHealth Predicts the Token Cost of Agentic Code Review: A Pre-Registered Study of 465 Agent-Authored Pull Requests
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed TurnCost: Turns, Not Observations: Cache-Aware Loop Engineering for Coding Agents
- Hardware and Architecture
- Information Systems
- Artificial Intelligence
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed Maximising Transparency in Context Compaction Improves Retention of Facts and Commitments for Lightweight Coding Agents
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed (When) Should a Software Factory be Lights-off? Benchmarking Code Review Agents on Bug Detection
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed TestSuiteBench: Coding Agents Struggle to Generate Strong Test Suites from Specifications
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed Does an Agent's History Tell You When Compaction Will Hurt? A Modest, Bounded Effect on the TRACE Paired-Replay Corpus
- Software
- Information Systems
- Computer Networks and Communications
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed Curating Always-Loaded Context for LLM Agents: A Capacitated Assortment Model with Censored Feedback
- Management Science and Operations Research
- Computer Networks and Communications
- Management Information Systems
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed Where Do the Tokens Go? Understanding and Reducing Costs in LLM Agents for Vulnerability Discovery
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed When Sub-Agents Work in Parallel: The Promises and Pitfalls of Dynamic Concurrency in Long-Horizon Coding Tasks
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed A Case Study in Assuring AI-Written Software
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed What Does a Harness Buy? Tokens, Mostly
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed Discriminating Fixture Coverage in Agent-Infrastructure Verification Suites
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed When Does a Second Model Help? Cross-Model Review in LLM Verification
- Information Systems and Management
- Information Systems
- Artificial Intelligence
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed AutoCompact: Learning When to Compact Context in Long-Horizon Coding Agents
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed Can Terminal Agents Trust Their Own Verification? Diagnosing and Improving Self-Verification
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed FOCUS: Training-Free Decision-Preserving Context Compression for LLM Agents
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed TokenCast: Forecasting Token Consumption During LLM Agent Execution
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed Analyzing and Mitigating Cost-Inefficient Behaviors in Coding Agents
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed Harness Tokenomics: A Router for the Enterprise Agentic Control Plane
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed Total Cost of Agency: Exact Attribution of Memory Injection Cost in Multi-Agent LLM Workflows
- Information Systems and Management
- Artificial Intelligence
- Materials Chemistry
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed An Empirical Study of Harness Design for Coding Agents
- Information Systems
- Artificial Intelligence
- Computer Networks and Communications
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed Quantifying Overclaiming Propensity in Frontier LLM Agents
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed Subagents vs Agent Skills: Executing Reusable Knowledge for Long-Horizon Agentic Tasks
- Artificial Intelligence
- Computer Vision and Pattern Recognition
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed Protocol-Preserving Context Trimming for Agentic Workflows: Benefits, Failure Regimes, and Budget Guardrails
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed Governed Human-AI Prioritization Under Uncertainty: Adaptive Estimation and Dependency-Constrained Portfolio Selection
- Artificial Intelligence
- Safety Research
- Management Science and Operations Research
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed The widening evaluation gap in medical large language model research 2023 to 2026
- Review published
- Preprint server
- arXiv
-
Evgenii Arsentev reviewed Reproducibility in the Age of Agentic AI: Context Engineering at the Timescale of a Codebase
- Review published
- Preprint server
- arXiv