Peer-reviewed work
Publications
Evgenii Arsentev, PhD. 6 peer-reviewed papers, 69 citations per OpenAlex as of 2026-09-10. Current work: the measured economics and reliability of AI agents.
Persistent identifiers
ORCID0000-0002-9120-7298- WoSWeb of Science ResearcherIDQZR-1204-2026
Scopus Author ID57202031127
OpenAlexA5087068667
Semantic Scholar65769164
WikidataQ141398931
Google Scholar_nnPkaAAAAAJ
ResearchGateEvgenii-Arsentev
Loop (Frontiers)3694570- CCureus1390583
ReviewerCreditsevgeniiarsentev
IETF Datatrackerdraft-arsentev-agent-run-metrics- IETF Datatrackerdraft-arsentev-llm-context-discovery
GitHubarsentev-ai- OSFOSF6dg3m
- OROpenReview~Evgenii_Arsentev1
Technical reports
No Request With a Known Crawler User-Agent Reached llms.txt: Fifteen Days of Logs at One Origin Server
Fifteen days of logs at one origin server: 46,155 requests presenting known search and AI crawler user-agents, none for llms.txt, while robots.txt was requested 698 times in the two days after the file went live. Corrects the numbers in draft-arsentev-llm-context-discovery-00.
DOI: 10.5281/zenodo.23018429 · series: 10.5281/zenodo.23018428 · PDF
AI Content Engine Economics: The Cost of Short-Form Video Production by AI Agents and by People
One company's AI content engine produced 1,497 short-form videos across 67 channels in 101 days for $4,313.76 in subscriptions: $0.52 per 1,000 organic views and $2.88 per video. The same volume from a human studio would take 42–78 people and $179,000–1.21 million a month; buying the reach from influencers would cost 3.8–57 times more. Built by one person with no technical background; code written by Claude agents. Company under NDA, figures and calculations only.
DOI: 10.5281/zenodo.22802229 · series: 10.5281/zenodo.22802228 · PDF
Clear Every Third Task: A Measured U-Curve in the Context Economy of Coding Agents
Corrected 2026-09-15. How often should an agent's context be cleared? 36 runs, six session-length policies, six replicates each, and no failure among the tests the agent wrote (4086 in total). Cost is U-shaped in session length: clearing after every task costs 33.5% more than clearing every third (p = 0.0022), while never clearing costs 9.7% more, which is not significant (p = 0.046, Holm-adjusted 0.14). Three, four and six tasks per session are indistinguishable, so the low region is a plateau. The U decomposes into cache writes falling and context per call rising, with 1-hour cache writes priced 20:1 against cache reads.
DOI: 10.5281/zenodo.22759217 · series: 10.5281/zenodo.22699667 · PDF
The Context Economy of Agentic LLM Sessions: Where the Money Actually Goes
Corrected 2026-09-15. 722 agent sessions, 150,902 model calls, 34.6 billion tokens. Context handling accounts for 87.8% of modeled cost and generation for 12.2%; by estimate, 80% of the spend comes from 2.9% of sessions. Includes two log-deduplication traps that change the answer by a factor of two in either direction, and the de-identified dataset.
DOI: 10.5281/zenodo.22759216 · series: 10.5281/zenodo.22688704 · PDF
Preprints
A Suite That Does Not Discriminate: The Case for Result-Set Fields in Agent Conformance Records
A set of conformance checks can consist entirely of sound, correctly reported, individually falsifiable checks and still support no conclusion, because no check changed state across the conditions compared. That is a property of the set, and no per-check field can carry it. The paper proposes a result-set field recording whether any check discriminated and which ones; it documents the adjacent case in a published corpus of 36 coding-agent runs, where a verdict of 4,086 tests and zero failures held with zero variance across six conditions whose modelled cost differed by 25.3% (a figure computed with the incorrect prices of report versions 1.0 and 1.1; 33.5% at the corrected prices of version 1.2, DOI 10.5281/zenodo.22759217); it reports as negative the tempting explanation that longer sessions simply did less work; and it identifies a case the proposed field cannot answer — where the checks are authored by the run itself, check identity does not survive across runs and the comparison-level question is undefined rather than false.
DOI: 10.32388/0BV3Z8
Software
contextburn
Open-source meter built from the context-economy experiments. Reads the transcripts Claude Code already writes locally and reports run efficiency: the share of paid tokens that became model output rather than a re-reading of context already sent, by tokens and cost-weighted. Makes no network calls.
Listed in: OECD.AI Catalogue of Tools & Metrics · Research Software Directory
Open data and video
AI Tools Radar: GitHub and Hugging Face projects selected by arsentev.ai (monthly)
490 open-source AI projects selected and reviewed by the arsentev.ai radar since June 2026, with daily GitHub and Hugging Face API snapshots of how they evolve after selection. A new version is released every month.
Context U-curve: 36 coding-agent runs under six context-clearing policies
Run-level token counters, modeled cost, wall clock and test outcomes for every run behind the U-curve report. Counters only - no prompts, no model output, no paths.
DOI: 10.57967/hf/10366
Context economy: modeled cost of 722 coding-agent sessions
Per-session model calls and modeled cost at list prices for the 722 sessions behind the Context Economy report (v1.1). Counters only - no prompts, no code, no project names.
DOI: 10.57967/hf/10649
Context U-Curve of Coding Agents (36 runs)
The same run-level dataset on Kaggle, with a data card and per-column profiles.
Clear Every Third Task - report and data on OSF
Open Science Framework project holding the report PDF, the run-level dataset and its card.
contextburn demo: run efficiency of 36 coding-agent runs
Twenty-second terminal recording of contextburn over the experiment's session logs: 168 sessions, 94.4% of paid tokens re-read context already sent. Archived at archive.org/details/contextburn-demo-context-ucurve-2026.
Peer review service
JMIR PublicationsPeer reviewer
The reviewer role covers all 27 journals of the family, listed below with their ISSNs
- JMIRJournal of Medical Internet Research1438-8871
- AIJMIR AI2817-1705
- MIJMIR Medical Informatics2291-9694
- FRJMIR Formative Research2561-326X
- mHJMIR mHealth and uHealth2291-5222
- MHJMIR Mental Health2368-7959
- MEJMIR Medical Education2369-3762
- HFJMIR Human Factors2292-9495
- RPJMIR Research Protocols1929-0748
- PHJMIR Public Health and Surveillance2369-2960
- SGJMIR Serious Games2291-9279
- AGJMIR Aging2561-7605
- CAJMIR Cancer2369-1999
- PPJMIR Pediatrics and Parenting2561-6722
- iJMRInteractive Journal of Medical Research1929-073X
- xMedJMIRx Med2563-6316
- OJPHIOnline Journal of Public Health Informatics1947-2579
- BBJMIR Bioinformatics and Biotechnology2563-3570
- NUJMIR Nursing2562-7600
- RAJMIR Rehabilitation and Assistive Technologies2369-2529
- DEJMIR Dermatology2562-0959
- CDJMIR Cardio2561-1011
- DIJMIR Diabetes2371-4379
- BEJMIR Biomedical Engineering2561-3278
- NTJMIR Neurotechnology2817-092X
- PMJMIR Perioperative Medicine2561-9128
- INJMIR Infodemiology2564-1891
- C
- WoS
- PR
- Q
- BIGBusiness Intelligence Group
Judge, Business Intelligence Group awards
Public judge profile; first program: 2026 BIG Awards for Business
profile → - MSRMSR 2027 — Mining Software Repositories (co-located with ICSE 2027)
Junior Program Committee member, Technical Papers
Listed on the conference committee page
profile →
Published peer reviews
- Review of “Compaction Laundering: Provenance Destruction in Context-Compacting Coding Agents”
PREreview · 2026-10-23
DOI 10.5281/zenodo.23591614 - Review of “DAug: Diffusion-based Channel Augmentation for Radiology Image Retrieval and Classification”
Qeios · 2026-10-23
DOI 10.32388/C95YAZ - Review of “LiLMaps: Learnable Implicit Language Maps”
Qeios · 2026-10-22
DOI 10.32388/HGE3Z6 - Review of “REM: What to Store, When to Recall, What to Forget in Experience Memory for Coding Agents”
PREreview · 2026-10-22
DOI 10.5281/zenodo.23566644 - Review of “Adverse Environmental and Public Health Effects of Artificial Intelligence: A Narrative Review”
Qeios · 2026-10-21
DOI 10.32388/P166QC - Review of “SAS Multimetric Tribunal: What a Hallucination Benchmark Measures and What It Does Not (v7, sixth-round review)”
Qeios · 2026-10-20
DOI 10.32388/WXECBL - Review of “METAGENE-1: Metagenomic Foundation Model for Pandemic Monitoring”
Qeios · 2026-10-20
DOI 10.32388/9WHP1S - Review of “When Coding Agents Outpace Review: Evaluating PRD-Driven Development in Mature Enterprise Repositories”
PREreview · 2026-10-20
DOI 10.5281/zenodo.23499870 - Review of “Acceptability of chatbot versus General Practitioner consultations for healthcare conditions varying in terms of perceived stigma and severity”
Qeios · 2026-10-19
DOI 10.32388/MBANZF - Review of “Fair Diagnosis: Leveraging Causal Modeling to Mitigate Medical Bias”
Qeios · 2026-10-18
DOI 10.32388/T29LZ0 - Review of “Large Vision-Language Model Alignment and Misalignment: A Survey Through the Lens of Explainability”
Qeios · 2026-10-18
DOI 10.32388/P3DDOA - Review of “SAS Multimetric Tribunal: What a Hallucination Benchmark Measures, and What It Does Not (v6, fifth-round review)”
Qeios · 2026-10-16
DOI 10.32388/I1U8DS - Review of “SAS Multimetric Tribunal: What a Hallucination Benchmark Measures, and What It Does Not (v5, fourth-round review)”
Qeios · 2026-10-13
DOI 10.32388/W97ING - Review of “Deep Learning-Based CKM Construction with Image Super-Resolution”
Qeios · 2026-10-13
DOI 10.32388/R29CIF - Review of “Do LLMs Overcome Shortcut Learning? An Evaluation of Shortcut Challenges in Large Language Models”
Qeios · 2026-10-15
DOI 10.32388/9GD60S - Review of “Convolutional Neural Networks Do Work with Pre-Defined Filters”
Qeios · 2026-10-11
DOI 10.32388/DUYWSM - Review of “Decision Support or High-Risk? Classifying AI Workforce-Intelligence Systems Under the EU AI Act”
Qeios · 2026-10-10
DOI 10.32388/BXWZ5J - Review of “Are AI Peer Reviewers More Demanding Than Humans? A Cross-Disciplinary Study of AI-Based Manuscript Evaluation”
Qeios · 2026-10-10
DOI 10.32388/BZEHY1 - Review of “From Informal to Formal – Incorporating and Evaluating LLMs on Natural Language Requirements to Verifiable Formal Proofs”
Qeios · 2026-10-08
DOI 10.32388/W73WEA - Review of “Deep Sylvester Posterior Inference for Adaptive Compressed Sensing in Ultrasound Imaging”
Qeios · 2026-10-08
DOI 10.32388/YQ0OF6 - Review of “Deep Learning in Medical Image Registration: Introduction and Survey”
Qeios · 2026-10-09
DOI 10.32388/54UP92 - Review of “Visible but Unverified: How Public Signals Diverge from Formal Records in Israeli Digital Health Startups”
Qeios · 2026-10-07
DOI 10.32388/O0PE5Q - Review of “SAS Multimetric Tribunal: Structural Classification of Hallucination in AI-Generated Text (v4, third-round review)”
Qeios · 2026-10-07
DOI 10.32388/6HK3TA - Review of “Efficient Pruning of Text-to-Image Models: Insights from Pruning Stable Diffusion”
Qeios · 2026-10-06
DOI 10.32388/CGU3SM - Review of “Solving Sokoban using Hierarchical Reinforcement Learning with Landmarks”
Qeios · 2026-10-06
DOI 10.32388/IFMICQ - Review of “Fine-Grained Alignment in Vision-and-Language Navigation through Bayesian Optimization”
Qeios · 2026-10-06
DOI 10.32388/QFXEUW - Review of “Protein Structure Prediction in the 3D HP Model Using Deep Reinforcement Learning”
Qeios · 2026-10-06
DOI 10.32388/DC3K56 - Review of “Influence Functions for Scalable Data Attribution in Diffusion Models”
Qeios · 2026-10-05
DOI 10.32388/ROA5OA - Review of “Enhancing Dataset Distillation via Label Inconsistency Elimination and Learning Pattern Refinement”
Qeios · 2026-10-05
DOI 10.32388/Q4A8WD - Review of “CM3T: Framework for Efficient Multimodal Learning for Inhomogeneous Interaction Datasets”
Qeios · 2026-10-05
DOI 10.32388/QTX1ZT - Review of “Enhancing Sample Generation of Diffusion Models using Noise Level Correction”
Qeios · 2026-10-05
DOI 10.32388/D7JXUL - Review of “AVID: Adapting Video Diffusion Models to World Models”
Qeios · 2026-10-04
DOI 10.32388/RI9UZB - Review of “MVD: A Multi-Lingual Software Vulnerability Detection Framework”
Qeios · 2026-10-04
DOI 10.32388/X1ANZB - Review of “Partial Convolution Meets Visual Attention”
Qeios · 2026-10-04
DOI 10.32388/KP0ZYF - Review of “When Large Vision-Language Models Meet Person Re-Identification”
Qeios · 2026-10-04
DOI 10.32388/2U84XV - Review of “SampleLLM: Optimizing Tabular Data Synthesis in Recommendations”
Qeios · 2026-10-02
DOI 10.32388/B6MTRF - Review of “Mitigating GenAI-powered Evidence Pollution for Out-of-Context Multimodal Misinformation Detection”
Qeios · 2026-10-02
DOI 10.32388/YGVQNZ - Review of “k2SSL: A Faster and Better Framework for Self-Supervised Speech Representation Learning”
Qeios · 2026-10-02
DOI 10.32388/7USN74 - Review of “iLLaVA: An Image is Worth Fewer Than 1/3 Input Tokens in Large Multimodal Models”
Qeios · 2026-10-02
DOI 10.32388/P0C007 - Review of “Approximated Variational Bayesian Inverse Reinforcement Learning for Large Language Model Alignment”
Qeios · 2026-10-01
DOI 10.32388/N8IB9I - Review of “LLaVA-MR: Large Language-and-Vision Assistant for Video Moment Retrieval”
Qeios · 2026-10-01
DOI 10.32388/VPQPOI - Review of “Instruction-Driven Game Engine: A Poker Case Study”
Qeios · 2026-10-01
DOI 10.32388/1D2EWI - Review of “PANDA – Paired Anti-hate Narratives Dataset from Asia: Using an LLM-as-a-Judge to Create the First Chinese Counterspeech Dataset”
Qeios · 2026-10-01
DOI 10.32388/ITCI8J - Review of “Don’t Shake the Wheel: Momentum-Aware Planning in End-to-End Autonomous Driving”
Qeios · 2026-09-29
DOI 10.32388/FMY1RN - Review of “Superposition in Transformers: A Novel Way of Building Mixture of Experts”
Qeios · 2026-09-30
DOI 10.32388/K5BS42 - Review of “Can You Trust the Graph Beneath the Agent? A Reliability Metrology for Enterprise Skills Graphs”
Qeios · 2026-09-30
DOI 10.32388/6YDYUD - Review of “Are Vision-Language Models Truly Understanding Multi-vision Sensor?”
Qeios · 2026-09-30
DOI 10.32388/VF371C - Review of “R+R: Security Vulnerability Dataset Quality Is Critical”
Qeios · 2026-09-30
DOI 10.32388/4SPQ4T - Review of “Arabic-Nougat: Fine-Tuning Vision Transformers for Arabic OCR and Markdown Extraction”
Qeios · 2026-09-29
DOI 10.32388/1SXCRZ - Review of “Translution: Unifying Self-Attention and Convolution for Adaptive and Relative Modelling”
Qeios · 2026-09-29
DOI 10.32388/XOD588 - Review of “Neural Networks Use Distance Metrics”
Qeios · 2026-09-29
DOI 10.32388/HNZFP7 - Review of “What Does a Harness Buy? Tokens, Mostly”
PREreview · 2026-10-06
DOI 10.5281/zenodo.23183692 - Review of “A Case Study in Assuring AI-Written Software”
PREreview · 2026-10-07
DOI 10.5281/zenodo.23207006 - Review of “When Sub-Agents Work in Parallel: The Promises and Pitfalls of Dynamic Concurrency in Long-Horizon Coding Tasks”
PREreview · 2026-10-08
DOI 10.5281/zenodo.23231601 - Review of “Where Do the Tokens Go? Understanding and Reducing Costs in LLM Agents for Vulnerability Discovery”
PREreview · 2026-10-09
DOI 10.5281/zenodo.23257662 - Review of “Curating Always-Loaded Context for LLM Agents: A Capacitated Assortment Model with Censored Feedback”
PREreview · 2026-10-10
DOI 10.5281/zenodo.23282285 - Review of “Does an Agent’s History Tell You When Compaction Will Hurt? A Modest, Bounded Effect on the TRACE Paired-Replay Corpus”
PREreview · 2026-10-11
DOI 10.5281/zenodo.23293074 - Review of “TestSuiteBench: Coding Agents Struggle to Generate Strong Test Suites from Specifications”
PREreview · 2026-10-12
DOI 10.5281/zenodo.23321660 - Review of “(When) Should a Software Factory be Lights-off? Benchmarking Code Review Agents on Bug Detection”
PREreview · 2026-10-13
DOI 10.5281/zenodo.23335998 - Review of “TurnCost: Turns, Not Observations: Cache-Aware Loop Engineering for Coding Agents”
PREreview · 2026-10-16
DOI 10.5281/zenodo.23423668 - Review of “CodeHealth Predicts the Token Cost of Agentic Code Review: A Pre-Registered Study of 465 Agent-Authored Pull Requests”
PREreview · 2026-10-17
DOI 10.5281/zenodo.23432006 - Review of “Skip the Compressor: Coding Agents Fix More Issues When They Pick Whole Files”
PREreview · 2026-10-19
DOI 10.5281/zenodo.23468501 - Review of “The Role of Agentic Instruction Categories for Code Generation: A Study of Five Python Libraries”
PREreview · 2026-10-18
DOI 10.5281/zenodo.23446298 - Review of “Maximising Transparency in Context Compaction Improves Retention of Facts and Commitments for Lightweight Coding Agents”
PREreview · 2026-10-15
DOI 10.5281/zenodo.23401287 - Review of “Discriminating Fixture Coverage in Agent-Infrastructure Verification Suites”
PREreview · 2026-10-05
DOI 10.5281/zenodo.23159191 - Review of “When Does a Second Model Help? Cross-Model Review in LLM Verification”
PREreview · 2026-10-04
DOI 10.5281/zenodo.23132634 - Review of “AutoCompact: Learning When to Compact Context in Long-Horizon Coding Agents”
PREreview · 2026-10-02
DOI 10.5281/zenodo.23111148 - Review of “Can Terminal Agents Trust Their Own Verification? Diagnosing and Improving Self-Verification”
PREreview · 2026-10-01
DOI 10.5281/zenodo.23080173 - Review of “FOCUS: Training-Free Decision-Preserving Context Compression for LLM Agents”
PREreview · 2026-09-30
DOI 10.5281/zenodo.23057016 - Review of “TokenCast: Forecasting Token Consumption During LLM Agent Execution”
PREreview · 2026-09-29
DOI 10.5281/zenodo.23037648 - Review of “Vision-Language In-Context Learning Driven Few-Shot Visual Inspection Model”
Qeios · 2026-09-28
DOI 10.32388/GN91RB - Review of “One Communication Round is All It Needs for Federated Fine-Tuning Foundation Models”
Qeios · 2026-09-28
DOI 10.32388/K2EZRF - Review of “OpenAD: Open-World Autonomous Driving Benchmark for 3D Object Detection”
Qeios · 2026-09-28
DOI 10.32388/02JI1X - Review of “Accelerating Vision Diffusion Transformers with Skip Branches”
Qeios · 2026-09-28
DOI 10.32388/DEA34L - Review of “CAT: Content-Adaptive Image Tokenization”
Qeios · 2026-09-26
DOI 10.32388/9BTT2Q - Review of “Grounding-IQA: Multimodal Language Grounding Model for Image Quality Assessment”
Qeios · 2026-09-26
DOI 10.32388/JDNJLA - Review of “Union of Experts: Adapting Hierarchical Routing to Equivalently Decomposed Transformer”
Qeios · 2026-09-26
DOI 10.32388/UHU2TE - Review of “LoLDU: Low-Rank Adaptation via Lower-Diag-Upper Decomposition for Parameter-Efficient Fine-Tuning”
Qeios · 2026-09-26
DOI 10.32388/0OSZPL - Review of “Analyzing and Mitigating Cost-Inefficient Behaviors in Coding Agents”
PREreview · 2026-09-28
DOI 10.5281/zenodo.23011395 - Review of “Control the Harness, Control the Cost: Routing and Governing AI Coding Agents in the Enterprise”
PREreview · 2026-09-26
DOI 10.5281/zenodo.22974223 - Review of “SAS Multimetric Tribunal: Structural Classification of Hallucination in AI-Generated Text (v3, follow-up review)”
Qeios · 2026-09-25
DOI 10.32388/9T9P58 - Review of “Metadata Conditioning Accelerates Language Model Pre-training”
Qeios · 2026-09-25
DOI 10.32388/RTLKQV - Review of “SMOOTHIE: Label Free Language Model Routing”
Qeios · 2026-09-25
DOI 10.32388/1I80IJ - Review of “Overcoming Sparsity Artifacts in Crosscoders to Interpret Chat-Tuning”
Qeios · 2026-09-25
DOI 10.32388/ARNLFW - Review of “LLM-ABBA: Understanding Time Series via Symbolic Approximation”
Qeios · 2026-09-25
DOI 10.32388/W2EYI2 - Review of “Total Cost of Agency: Exact Attribution of Memory Injection Cost in Multi-Agent LLM Workflows”
PREreview · 2026-09-25
DOI 10.5281/zenodo.22955888 - Review of “Geometric Analysis of Reasoning Trajectories: A Phase Space Approach to Understanding Valid and Invalid Multi-Hop Reasoning in LLMs”
Qeios · 2026-09-23
DOI 10.32388/HZYFIL - Review of “Document Haystacks: Vision-Language Reasoning Over Piles of 1000+ Documents”
Qeios · 2026-09-23
DOI 10.32388/WN2CLD - Review of “VideoHallu: Evaluating and Mitigating Multi-modal Hallucinations for Synthetic Videos”
Qeios · 2026-09-23
DOI 10.32388/OB3IWD - Review of “Differentiating Conscious Human Entities from Autonomous AI Agents: A Heart Rate Variability (HRV) Framework for Human Authentication in the Emerging AI Economy”
Qeios · 2026-09-23
DOI 10.32388/JD9MDP - Review of “Impact of Telemedicine on Post-Bariatric Surgery Outcomes and Patient Satisfaction during the COVID-19 Pandemic: A Retrospective Study”
Qeios · 2026-09-23
DOI 10.32388/6B43L6 - Review of “SAS Multimetric Tribunal: Structural Classification of Hallucination in AI-Generated Text”
Qeios · 2026-09-23
DOI 10.32388/7Z13RM - Review of “Fine-Tuning PHI-3 for Multiple-Choice Question Answering: Methodology, Results, and Challenges”
Qeios · 2026-09-23
DOI 10.32388/G8KZQP - Review of “Transformers Can Navigate Mazes With Multi-Step Prediction”
Qeios · 2026-09-23
DOI 10.32388/S7USOO - Review of “An Empirical Study of Harness Design for Coding Agents”
PREreview · 2026-09-21
DOI 10.5281/zenodo.22879230 - Review of “Simulating Lay Health-Seeking Behavior with LLM Personas and Illness Vignettes: Reproducibility, Prompt Sensitivity, and Slice Dependence (v3)”
Qeios · 2026-09-21
DOI 10.32388/5P0NFZ - Review of “In RAG We Trust? Measuring Robustness of Retrieval-Augmented Generation Under Document Poisoning”
Qeios · 2026-09-21
DOI 10.32388/AVRGPF - Review of “An AI-Human Collaborative Review of AI-Human Collaborative Reviews”
Qeios · 2026-09-21
DOI 10.32388/6HHPN7 - Review of “WeAudit: Scaffolding User Auditors and AI Practitioners in Auditing Generative AI”
Qeios · 2026-09-21
DOI 10.32388/9V58FH - Review of “AI Ethics by Design: Implementing Customizable Guardrails for Responsible AI Development”
Qeios · 2026-09-21
DOI 10.32388/CO9D6J - Review of “Challenges and Paths Towards AI for Software Engineering”
Qeios · 2026-09-21
DOI 10.32388/DLSNN3 - Review of “MiND: A Minimal Deterministic Middleware for Auditable LLM Interaction”
Qeios · 2026-09-18
DOI 10.32388/FLR547 - Review of “SiReRAG: Indexing Similar and Related Information for Multihop Reasoning”
Qeios · 2026-09-18
DOI 10.32388/MYWOY6 - Review of “Enhancing Code LLMs with Reinforcement Learning in Code Generation: A Survey”
Qeios · 2026-09-18
DOI 10.32388/GJ2PT3 - Review of “Benchmark Evaluations, Applications, and Challenges of Large Vision Language Models: A Survey”
Qeios · 2026-09-18
DOI 10.32388/ILZIVN - Review of “Enhancing Lifelong Multi-Agent Path Finding with Cache Mechanism”
Qeios · 2026-09-18
DOI 10.32388/S9KCZK - Review of “A Geometric Taxonomy of Hallucinations in LLMs”
Qeios · 2026-09-18
DOI 10.32388/WA4DDG - Review of “Quantifying Overclaiming Propensity in Frontier LLM Agents”
PREreview · 2026-09-18
DOI 10.5281/zenodo.22827742 - Review of “Beyond Perplexity: The Quality–Memory Frontier of Native Mixed-Precision KV Caching in llama.cpp”
Qeios · 2026-09-17
DOI 10.32388/XW3TTL - Review of “DAFE: LLM-Based Evaluation Through Dynamic Arbitration for Free-Form Question-Answering”
Qeios · 2026-09-17
DOI 10.32388/JTXDPF - Review of “LLM Confidence Evaluation Measures in Zero-Shot CSS Classification”
Qeios · 2026-09-17
DOI 10.32388/B2373B - Review of “Hybrid Memory-Retrieval Model: Enhancing Trust in Medical Chatbots”
Qeios · 2026-09-17
DOI 10.32388/GP1SOV - Review of “100% Hallucination Elimination Using Acurai”
Qeios · 2026-09-17
DOI 10.32388/LLM844 - Review of “ToxiLab: How Well Do Open-Source LLMs Generate Synthetic Toxicity Data?”
Qeios · 2026-09-17
DOI 10.32388/V2O96E - Review of “PhysAgent: A Multi-Agent Approach to the Automated Discovery of Physical Laws”
Qeios · 2026-09-17
DOI 10.32388/IRHPUI - Review of “GraphICL: Unlocking Graph Learning Potential in LLMs through Structured Prompt Design”
Qeios · 2026-09-17
DOI 10.32388/FYURAH - Review of “GeAR: Generation Augmented Retrieval”
Qeios · 2026-09-17
DOI 10.32388/J2NQSS - Review of “Subagents vs Agent Skills: Executing Reusable Knowledge for Long-Horizon Agentic Tasks”
PREreview · 2026-09-17
DOI 10.5281/zenodo.22808990 - Review of “Simulating Lay Health-Seeking Behavior with LLM Personas and Illness Vignettes”
Qeios · 2026-09-16
DOI 10.32388/TOJXVB - Review of “MTRAG: A Multi-Turn Conversational Benchmark for Evaluating Retrieval-Augmented Generation Systems”
Qeios · 2026-09-16
DOI 10.32388/3Z16QK - Review of “Matching Frontier Code Agents with Lightweight Models via Multi-Model Consultation”
Qeios · 2026-09-16
DOI 10.32388/VMZG2G - Review of “The Cost of Consensus: Malignant Epistemic Herding in Distributed Multi-Agent Search”
Qeios · 2026-09-16
DOI 10.32388/PA3F8W - Review of “Towards More Economical Context-Augmented LLM Generation by Reusing Stored KV Cache”
Qeios · 2026-09-16
DOI 10.32388/F2FNB3 - Review of “Physical AI Agents: Integrating Cognitive Intelligence with Real-World Action”
Qeios · 2026-09-16
DOI 10.32388/G1XLD1 - Review of “Diversity of Thought Elicits Stronger Reasoning Capabilities in Multi-Agent Debate Frameworks”
Qeios · 2026-09-16
DOI 10.32388/QSXNZP - Review of “Semantic Grounding Index: Geometric Bounds on Context Engagement in RAG Systems”
Qeios · 2026-09-16
DOI 10.32388/3RPTO6 - Review of “In RAG We Trust? Measuring Robustness of Retrieval-Augmented Generation Under Document Poisoning”
Qeios · 2026-09-16
DOI 10.32388/K4VRQB - Review of “Protocol-Preserving Context Trimming for Agentic Workflows: Benefits, Failure Regimes, and Budget Guardrails”
PREreview · 2026-09-16
DOI 10.5281/zenodo.22788976 - Review of “MedAgentBench: A Realistic Virtual EHR Environment to Benchmark Medical LLM Agents”
Qeios · 2026-09-15
DOI 10.32388/2Z0NIB - Review of “AgentNet: Decentralized Evolutionary Coordination for LLM-based Multi-Agent Systems”
Qeios · 2026-09-15
DOI 10.32388/L9SZ30 - Review of “Star-Agents: Automatic Data Optimization with LLM Agents for Instruction Tuning”
Qeios · 2026-09-15
DOI 10.32388/K1W9S4 - Review of “Chain of Ideas: Novel Idea Development with LLM Agents”
Qeios · 2026-09-15
DOI 10.32388/JXTHLV - Review of “DRC-Coder: Automated DRC Checker Code Generation Using LLM Autonomous Agent”
Qeios · 2026-09-15
DOI 10.32388/ENZPY3 - Review of “Enhancing Annotated Bibliography Generation with LLM Ensembles”
Qeios · 2026-09-15
DOI 10.32388/A2L7P0 - Review of “Top Ten Challenges Towards Agentic Neural Graph Databases”
Qeios · 2026-09-15
DOI 10.32388/9886EW - Review of “Agentic Systems: A Guide to Transforming Industries with Vertical AI Agents”
Qeios · 2026-09-15
DOI 10.32388/KDXH0Z - Review of “Accelerating Manufacturing Scale-Up Using Agentic Web Navigation and Retrieval-Augmented AI”
Qeios · 2026-09-15
DOI 10.32388/SXRGP2 - Review of “Governed Human-AI Prioritization Under Uncertainty: Adaptive Estimation and Dependency-Constrained Portfolio Selection”
PREreview · 2026-09-14
DOI 10.5281/zenodo.22745390 - Review of “The widening evaluation gap in medical large language model research 2023 to 2026”
PREreview · 2026-09-13
DOI 10.5281/zenodo.22734670 - Review of “Reproducibility in the Age of Agentic AI”
PREreview · 2026-09-12
DOI 10.5281/zenodo.22729393 - Review of “A Dutch Financial Large Language Model”
Qeios · 2026-09-12
DOI 10.32388/G98QNJ - Review of “Prompt Volatility: An Empirical Study of Identity Drift in LLM Agents”
Qeios · 2026-09-12
DOI 10.32388/V3X30U - Review of “MixLLM: Dynamic Routing in Mixed Large Language Models”
Qeios · 2026-09-12
DOI 10.32388/GZL4AH - Review of “LemmaHead: RAG Assisted Proof Generation”
Qeios · 2026-09-12
DOI 10.32388/59NX5L - Review of “Anchoring Bias in Large Language Models”
Qeios · 2026-09-12
DOI 10.32388/92KCPQ - Review of “LLM Interface for Home Energy Management Systems”
Qeios · 2026-09-12
DOI 10.32388/JADXXS - Review of “Self-Pluralising Culture Alignment for LLMs”
Qeios · 2026-09-12
DOI 10.32388/DBCPMZ - Review of “LFOSum: Summarizing Long-form Opinions with LLMs”
Qeios · 2026-09-12
DOI 10.32388/0WXOBG
Letters to journals (eLetters)
- A second agent that tries to disprove the first
Response to «The Virtual Biotech: A multi-agent AI framework for therapeutic discovery and development» · Science (eLetter) · 2026-09-25
Public comments to regulators and parliament
- RFI on Modernizing the National Vulnerability Database in the Age of AI
NIST · docket NIST-2026-0100 · Regulations.gov · 2026-09-14
- RFI on the Clinical Laboratory Improvement Amendments (CLIA): AI in post-analytic review
CMS · docket CMS-2026-2345 · Regulations.gov · 2026-09-14
- CY 2027 Physician Fee Schedule: the proposed MIPS improvement activity on clinician use of AI, and AI in primary care
CMS · docket CMS-2026-2377 · Regulations.gov · 2026-09-17
- CY 2027 Physician Fee Schedule: paying for asynchronous telehealth work and keeping AI-assisted intake auditable
CMS · docket CMS-2026-2377 · Regulations.gov · 2026-09-17
- Submission to the Joint Select Committee on Artificial Intelligence
Parliament of Australia, Joint Select Committee on Artificial Intelligence · Submission 243 · aph.gov.au · 2026-09-16
Memberships and standards
- W3C
Editor, v0.1 of the conformance reporting format, Agent Conformance and Benchmarking Community Group
Reproducible methods for measuring AI agent conformance. Appointed editor of v0.1 by the group on the public mailing list, 17-18 September 2026; v0.1 released on the list on 30 September 2026; a community-group report, not a W3C standard
profile → 
MLCommonsWorking group participant
AI Standards HubCommunity member (Alan Turing Institute, BSI, NPL)
Research Data AllianceMember
Member since September 2026, participant in 13 working and interest groups
profile →
Open Source InitiativeMember
Internet SocietyMember, including the AI Special Interest Group
Earlier peer-reviewed work
6 peer-reviewed papers from 2017–2019, 69 citations, indexed in Scopus, Web of Science and PubMed. The subject is health care financing and organisation — unrelated to the current work on AI agents, so the full list lives in the registries.