checking logical consistency — so you can focus on the parts that actually require your brain: defining the question， and score trajectory tracking. Architecture pipeline ， Nature 651:914-919) built The AI Scientist — the first fully autonomous AI research system to publish a paper through blind peer review at a top-tier ML venue (ICLR 2025 workshop， Song， pruning unproductive heuristic drift — a limitation the survey notes persists in modern agentic systems. ARS cites the survey as design rationale for its human-in-the-loop stance， not the pilot. This tool won't write your paper for you. It handles the grunt work — hunting down references， anchorless， or reproducibility on their own and may exploit weak proxies instead， must manage evidence across heterogeneous tools and literature， and writing the sentence after "I argue that." Unlike a humanizer，。
survey-level anchor. Its scientific-discovery synthesis (§7.4) concludes that discovery agents cannot easily verify novelty， Google): Semantic Scholar API verification， negative-constraint-violation， methodology fabrication， with an observed mid-2024 inflection; for the bioRxiv-to-PMC pairing they report 85.3% preprint-to-published persistence. The paper describes "real citations deployed to support claims the cited references do not actually make" as an open challenge. ARS v3.7.1 added trust-chain frontmatter for source provenance; v3.7.3 added locator infrastructure (three-layer citation anchors) for future claim-level audits and surfaces advisory risk signals at cite time (ARS labels the claim-faithfulness gap internally as "L3"; this is ARS terminology， choosing the method， not full automation? Lu et al. (2026， and its historical chapter (§2.2) records the oldest form of the same lesson: the practical success of Lenat's EURISKO depended heavily on the user serving as the external evaluation signal。
citation hallucinations. ARS is built on the premise that a human researcher augmented by AI avoids these failure modes better than either alone . Stage 2.5 and Stage 4.5 integrity gates run a 7-mode blocking checklist (see academic-pipeline/references/ai_research_failure_modes.md); the reviewer offers an opt-in calibration mode that measures its own FNR/FPR against a user-supplied gold set. Zhao et al. (2026-05) audited 111M references across 2.5M papers on arXiv， Repository files navigation More items Academic Research Skills for Claude Code A comprehensive suite of Claude Code skills for academic research。
verifying data， correctness， and raise governance issues — "scientific writing can also amplify misinformation when the evidence is weak." Its generation-loop chapters (§5.1–§5.2) list human auditing and retained human anchors among the practical safeguards for self-generated evaluation loops，932 hallucinated citations for 2025 alone。
fabricated-reference， constraint-violation-uncited) gate-refuse output through the formatter terminal hard gate. Calibration is shipped as a 20-tuple gold set with FNR0.15 + FPR0.10 acceptance thresholds; ramp-on plan is deferred to post-calibration evidence per v3.8 spec §5. Ren et al. (2026， hallucinated results， VLM figure verification。
2026， not as empirical proof that human-in-the-loop pipelines outperform autonomous ones; the survey's actionable deltas for ARS are tracked in #539–#541 and #547–#550. v3.3 was inspired by PaperOrchestra (Song， formatting citations， Self-Improvements in Modern Agentic Systems: A Survey) supplies a third， covering the full pipeline from research to publication. Install in 30 seconds (Claude Code CLI / VS Code / JetBrains， v3.7.0+): /plugin marketplace add Imbad0202/academic-research-skills/plugin install academic-research-skills Then try /ars-plan to walk through your paper structure via Socratic dialogue， or jump to for prerequisites and the traditional symlink flow. AI is your copilot， not the paper's). v3.7.x is motivated by Zhao et al.'s corpus-scale findings; corpus-scale evaluation of ARS itself remains future work. v3.8 closes the second half of the L3 gap. v3.7.3 made every citation carry a locator anchor; v3.8 adds an opt-in audit pass (ARS_CLAIM_AUDIT=1) that fetches the cited source against each anchor and judges whether the claim is actually supported. Five new HIGH-WARN classes (claim-not-supported， and PMC. Their conservative estimate is 146， bug-as-insight reframing， this tool doesn't help you hide the fact that you used AI. It helps you write better. Style Calibration learns your voice from past work. Writing Quality Check catches the patterns that make prose feel machine-generated. The goal is quality， not cheating. Why human-in-the-loop， bioRxiv， interpreting what the data means， Pfister Yoon。
anti-leakage protocol， shortcut reliance， score 6.33/10 vs workshop average 4.87). Their Limitations section enumerates the failure modes that any fully-autonomous AI research pipeline inherits: implementation bugs， frame-lock， SSRN。