Literature

202 sources · 8 clusters · Live from litboard

Stack and sort research by cluster or see the gaps along the spine. Eight clusters across the full source inventory — filter by spine, tag, status, and outreach link. Click any tile for the full annotation.

Ait · Batzner · Hovy, D · Hanna · Grieve · Hovy, E · Parrish · Tsvetkov

A causal chain from human cognitive drive to corpus structure to model architecture to deployment harm, traced across six decades and four disciplines. FVE-1 and DIP positioned against BBQ, SYCON Bench, Batzner's missing HitL, Hanna's accountability framework, and Hovy's five-source taxonomy.

DIP Protocol is pre-operational pending expert consultation. See the DIP page for outreach details. DIP page →

Frenkel-Brunswik · Labov · Barthes · Staw · Fauconnier · Heyes · Masicampo · Fisher · Storm · Hubinger · Ouyang · Wei · Zhang · Hase · Kalra · FVE-1

Sixteen sources across seven decades converging on the same structural claim from cognitive psychology, narrative theory, organizational behavior, evolutionary cognition, internet cognition research, and LLM alignment. The drive toward closure is not a training artifact. It is the inherited grammar of literate culture's entire output, filtered through the cognitive architecture of the humans who produced it.

Sociology, linguistics, and computation walk into a bar...

A cluster map tracing the Gricean violation causal chain — from cooperative principles theory through corpus linguistics to accommodation research and non-Anglophone pragmatics. The claim: language models were theoretically grounded in Gricean cooperation but trained on its systematic violation. That incoherence is architecturally stable. Non-Anglophone speakers encounter it as a breakdown in the operative pragmatic norm without a name for what went wrong.

Gharib · Hussain · Auerbach · Hopfield · Mehta · Lin · Friston · Millidge · Adamczyk · Xie · IAIFI · FVE-1

Sixteen sources mapping physics, statistical mechanics, fluid dynamics, and computational neuroscience onto the behavioral substrate of FVE-1. Vortex ring formation mechanics as the physical model for BOWL and DRILL. The isomorphism is not decorative — Hopfield showed neural networks are physical systems in 1982, Mehta and Schwab showed the connection is exact in 2014. The forensic instruments read the residue. The physics cluster names the mechanism.

FVE-1 · BSR · BOWL · DRILL · CAPTURE — Structured Confound Inventory

A structured inventory of confounds that could independently distort the observed signal, the chosen order parameters, or the interpretation that behavioral signals reflect structured transitions rather than prompt artifacts, measurement artifacts, or investigator-side construction effects. Current literature mapped against the physics isomorphism claim.

FVE-1 Forensic Instrument vs. Cheng et al.

FVE-1 positioned against Cheng et al.'s work on model behavior. Where the accounts converge, where they diverge, and what each can see that the other cannot.

FVE-1 Behavioral Instrument vs. Wong Epistemological Framework

FVE-1 positioned against Wong's epistemological framework. The convergence table maps shared claims. The gap analysis identifies what each instrument can reach and what it cannot.

Frenkel-Brunswik · Labov · Wong · Cheng · FVE-1

The foundational through-line. A compulsion named as a human cognitive drive in 1949 traces through narrative grammar, corpus linguistics, and NLP into the forensic record of a language model session in 2026. Two domains, seven decades, one structural claim.

FVE-1 / Loss Landscape Framework v14 vs. Mechanistic Interpretability

How the FVE-1 behavioral instrument and the loss landscape framework map onto Neel Nanda's mechanistic interpretability research. Two programs working from opposite ends of the same architecture — one inside the weights, one reading the forensic residue from outside.