Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 48 min ago
PreprintCC BY 4.0arXiv

ForkSCOPE: Charting the Agentic Garden of Forking Paths

  • Arjun Balaji
  • Batuhan Duru Yeltekin
  • Tian Zheng
Open at sourcePDFDOICite

Abstract

Even with a fixed dataset and research question, data analysis involves many defensible decisions. Understanding how these choices influence the results is scientifically important but remains challenging. Crowdsourcing and agentic AI can generate hundreds of end-to-end analyses, but scaling generation alone can create a processing bottleneck and an analytic ``black hole.'' A common workaround is to impose a shared fixed decision taxonomy, which can limit insight and understate uncertainty. We present ForkSCOPE, a human-AI collaboration framework that induces structure bottom-up from the code corpus of end-to-end analyses, without a taxonomy fixed before or after generation, so the organization and evaluation of the garden can scale with the corpus. ForkSCOPE surfaces the charted garden of forking paths through a human-AI collaboration pipeline and an evidence-linked interactive viewer for steering and verification: it spotlights organically identified forks and structures and produces a derived taxonomy and decision map compatible with existing multiverse tools.

Related

  1. A counterexample to the stable forking conjecture
    James Freitag, Scott Mutchnik
    arXivPreprintCC BY 4.0Full textAlgebra & number theoryLogic & foundationsarXiv
  2. Forking in monadically NIP theories
    Artem Chernikov
    arXivPreprintarXiv licenceFull textLogic & foundationsarXiv
  3. Kim-forking for hyperimaginaries in nsop1 theories
    Yvon Bossut
    HALPreprintCC BY 4.0Full textLogic & foundationsHAL
  4. Forking and invariant types in monadic NIP theories
    Michael C. Laskowski
    arXivPreprintCC BY 4.0Full textLogic & foundationsarXiv
  5. Reducing stable forking dependence to finitely many pregeometries
    Scott Mutchnik
    arXivPreprintCC BY 4.0Full textCombinatorics & discrete mathematicsLogic & foundationsarXiv

Cite

BibTeXRIS
BibTeX
@misc{balaji2026forkscope,
  title = {{ForkSCOPE: Charting the Agentic Garden of Forking Paths}},
  author = {Arjun Balaji and Batuhan Duru Yeltekin and Tian Zheng},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.12438},
  eprint = {2609.12438},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.12438},
  note = {CC BY 4.0},
}
RIS
TY  - UNPB
TI  - ForkSCOPE: Charting the Agentic Garden of Forking Paths
AU  - Arjun Balaji
AU  - Batuhan Duru Yeltekin
AU  - Tian Zheng
PY  - 2026
DA  - 2026/09/14
PB  - arXiv
DO  - 10.48550/arxiv.2609.12438
UR  - https://arxiv.org/abs/2609.12438
AB  - Even with a fixed dataset and research question, data analysis involves many defensible decisions. Understanding how these choices influence the results is scientifically important but remains challenging. Crowdsourcing and agentic AI can generate hundreds of end-to-end analyses, but scaling generation alone can create a processing bottleneck and an analytic ``black hole.'' A common workaround is to impose a shared fixed decision taxonomy, which can limit insight and understate uncertainty. We present ForkSCOPE, a human-AI collaboration framework that induces structure bottom-up from the code corpus of end-to-end analyses, without a taxonomy fixed before or after generation, so the organization and evaluation of the garden can scale with the corpus. ForkSCOPE surfaces the charted garden of forking paths through a human-AI collaboration pipeline and an evidence-linked interactive viewer for steering and verification: it spotlights organically identified forks and structures and produces a derived taxonomy and decision map compatible with existing multiverse tools.
ER  -