Latent Relatedness: Attention as Relational

Inference and the Unintended Proﬁle Engine

Andres G.

Independent Research — PaxLabs Inc.

Pembroke Pines, FL, USA

Correspondence: legal@paxeer.app

Abstract—Self-attention is commonly described as weighted

aggregation over a context window: a mechanism for selecting

and blending relevant tokens. This account is not wrong, but

it understates what the mechanism makes possible. Attention

computes a context-dependent ﬁeld of pairwise semantic afﬁnities

over all tokens in context; composed across heads and layers,

this ﬁeld renders relational structure—including latent struc-

ture named by no single token—available to the rest of the

network. We introduce and formally deﬁne latent relatedness:

the phenomenon whereby a concept that is never mentioned, but

toward which a set of scattered signals jointly points, becomes

a high-probability continuation under a competent model. We

are careful to locate the inference correctly. Attention routes and

mixes relational information; the feed-forward layers—which be-

have as key–value associative memories that promote concepts in

vocabulary space—together with the unembedding condense that

information into an output distribution. Attributing the recovery

to attention alone is a mechanistic overstatement, and we correct

it. We then draw the consequence for user modeling, and separate

two claims that the literature and the popular framing tend to

conﬂate. The capacity to infer unstated attributes from scattered

text is an unavoidable property of capable models and is already

empirically demonstrated; the operationalization of that capacity

into a persistent, sharpening user proﬁle is not automatic and

requires external scaffolding—durable memory and periodic re-

analysis. This distinction dissolves an apparent paradox (proﬁling

as simultaneously a “free” side effect and an engineered ﬂywheel)

and re-frames the governance problem: the exposure is not a

stored dossier awaiting deletion but a reconstructable inference

that standard data-protection obligations still reach. We situate

the argument as the third of three papers on deployed-agent

architecture, and we state explicitly what is demonstrated, what

is argued, and what remains conjecture.

Index Terms—attention mechanisms, transformer inter-

pretability, distributional semantics, in-context learning, attribute

inference, user proﬁling, privacy, large language models

I. INTRODUCTION

E begin with a probe the reader can run on any

competent language model. Present the following eight

tokens with no framing and ask what they have in common:

love, family, small-town pride, heartbreak, whiskey,

ﬁshing, baseball, trucks.

Models tend to answer, and to do so conﬁdently,

with country music, or rural Americana, or working-

class emotional storytelling. The exact phrasing varies; a

convergence toward a single latent theme is typical. Neither

“country” nor “music” appears in the input. The label is not

present; it is recoverable from the structure of what is present.

This paper is about what that recoverability is, what mech-

anism produces it, and what follows for any system that

accumulates a user’s utterances over time. We make three

moves. First, we give a corrected account of the mecha-

nism: attention constructs a relational ﬁeld of token afﬁni-

ties, but the recovery of an unnamed center is a property

of the whole network—attention plus feed-forward memory

plus unembedding—not of attention in isolation. Attributing

the inference to attention alone, as an intuitive shorthand

sometimes does, misplaces the computation and does not

survive contact with the interpretability literature. Second, we

deﬁne the phenomenon operationally as latent relatedness and

distinguish it from literal retrieval, deductive inference, and

surface co-occurrence, while being candid that at the level of

raw computation it is high-probability next-token continuation;

the contribution is the framing and its consequence, not the

discovery of a new operation. Third, we separate the capacity

for latent inference—unavoidable, and already demonstrated

empirically—from its operationalization into a durable user

proﬁle, which requires engineered scaffolding. That separation

is what lets us speak precisely about governance.

A. Contributions

1) A corrected mechanistic account that locates unnamed-

center recovery in the composition of attention (rela-

tional routing) and feed-forward key–value memories

(associative promotion), rather than in attention alone

(Section III).

2) A formal, falsiﬁable deﬁnition of latent relatedness, with

a control condition that separates it from surface co-

occurrence, and a replication protocol (Section IV).

3) A capacity/operationalization distinction that resolves

the apparent contradiction between “proﬁling is a free

side effect” and “proﬁling requires an engineered ﬂy-

wheel,” grounding the capacity claim in demonstrated

attribute-inference results (Sections V–VI).

4) A governance analysis corrected for data-protection re-

ality: inferred attributes are reconstructable, not unreg-

ulated, and remediation attaches to the stored signal

set and the inference-time behavior rather than to a

nonexistent proﬁle record (Section VII).

5) An explicit statement of scope separating what is demon-

strated, argued, and conjectured (Section VIII).

B. Placement in a triptych

This is the third of three technical reports on the architecture

of deployed agents. The ﬁrst argued that deployed capability is

mediated by the scaffolding around a model—typed represen-

tations, persistent memory, deterministic execution, veriﬁable

provenance [1]. The second argued that autoregressive models

are structurally blind to absence: what is not present in context

exerts no force on generation, so gaps must be reiﬁed into

explicit context objects before they can be acted upon [2].

This paper concerns an emergent property rather than an en-

gineering prescription or a structural limit: the relational ﬁeld

that attention constructs makes latent structure recoverable.

We return to how the three interlock in Section VI, and we

are careful there to keep the emergent claim distinct from the

engineered one.

II. BACKGROUND AND RELATED WORK

A. Distributional semantics and the geometry of meaning

The premise that a word’s meaning is characterized by

the company it keeps is the distributional hypothesis [4],

[5]. Its modern realization—count-based and predictive word

embeddings [6], [7]—places related terms in nearby regions of

a vector space and exhibits additive compositional structure,

so that a set of related words has a centroid lying near terms

denoting their shared theme. This is the setting in which

the intuitive picture of “a center that minimizes distance to

a cluster” is literally apt. We stress this lineage because,

as Section III argues, that picture is an intuition imported

from static embedding geometry; it is not a description of

the computation a transformer performs at generation time.

B. What attention computes

The transformer [3] replaced recurrence with self-attention,

in which each position forms a query that is scored against

every key and the resulting weights mix value vectors.

Analyses of trained attention ﬁnd heads that implement

interpretable, often relational, routing—syntactic dependen-

cies, coreference, delimiter tracking [9]—and probing work

shows that structural information such as syntax trees is

linearly recoverable from contextual representations [10].

The mechanistic-interpretability program formalizes attention

heads as information-movement operators composing across

layers [11], and identiﬁes induction heads as a concrete

circuit that copies and completes patterns across a context,

a mechanism tied to in-context learning [12].

Two ﬁndings from this literature are load-bearing for us.

First, feed-forward layers act as key–value memories: each

key correlates with input patterns and each value induces

a distribution over the vocabulary, with the layer’s output

a composition of these memories reﬁned through residual

connections into the ﬁnal output distribution [13]. Second, this

promotion of concepts happens in vocabulary space [14]. To-

gether they imply that the associative “recall” of an unnamed

concept is done substantially by the feed-forward memories

and the unembedding—not by attention, whose role is to

assemble the context-dependent representation those memories

act on.

C. In-context learning

Large models perform tasks speciﬁed only by their context,

without weight updates [15]. Latent relatedness is a special,

unsupervised case: the “task” is never stated, and the model

completes toward the theme that best organizes the context.

The induction-head account [12] is one mechanistic ingredient

of this broader competence.

D. Inference of personal attributes

The user-modeling consequence we develop is not hypo-

thetical. Staab et al. present the ﬁrst comprehensive study

of pretrained models inferring personal attributes from free

text, showing that current models recover attributes such as

location, income, and sex from unstructured Reddit posts at up

to 85% top-1 and 95% top-3 accuracy, at a small fraction of the

cost and time of human analysts, and they extend the threat to

conversational settings [16]. This is the empirical backbone of

our capacity claim (Section V): what we describe abstractly as

recovering a latent center is, applied to a person’s utterances,

demonstrated attribute inference. The point of departure for

this paper is not that models can infer attributes—that is

established—but the mechanism-level framing of why, and the

resulting governance analysis.

III. ATTENTION AS RELATIONAL INFERENCE

A. The standard account and its incompleteness

Self-attention is conventionally cast as a way to “focus on”

relevant context when forming each token’s representation.

The story is one of weighted aggregation: query–key matching

yields scores, softmax yields weights, and those weights blend

value vectors into the current state. This is accurate as far

as it goes, but it frames attention as a retrieval operation—

pulling in relevant context—rather than as a relational one

that constructs a ﬁeld of afﬁnities over the entire input.

The relational framing predicts a phenomenon the retrieval

framing does not: that structure never named by any token

can nonetheless become recoverable.

B. The afﬁnity ﬁeld, precisely

Let X ∈ R

n×d

be the residual-stream representation of n

positions. A single attention head forms

Q = XW

, K = XW

, V = XW

, (1)

and computes the row-stochastic afﬁnity matrix

A = softmax



⊤

√



, A

exp



⊤

√



′

exp



⊤

′

√



(2)

so each row A

i,:

is a probability distribution over positions:

a normalized map of position i’s semantic afﬁnity to every

other position. The head’s contribution to the stream is AV ,

and with H heads and residual updates,

← x

h=1





. (3)

The collection {A

(ℓ,h)

} over layers ℓ and heads h is what we

call the relational ﬁeld: a stack of weighted graphs over the

input in which membership in a densely inter-attending cluster

is an explicit, computed quantity. It is here—and only here—

that “how each token relates to every other token” lives as an

object the network can compute over.

C. Where the center is actually recovered

It is tempting to say that the ﬁeld “points to” the unnamed

center and that generation “samples from the densest region

of the ﬁeld.” That is a useful intuition and a misleading

mechanism. The afﬁnity ﬁeld does not, by itself, emit country

music. The recovery is completed downstream. A feed-forward

layer applies

FFN(x) = σ





, (4)

where, following [13], the rows of W

act as keys that ﬁre

on input patterns and the rows of W

act as values that

push probability mass toward particular vocabulary items; the

output is a composition of these memories, reﬁned through the

residual stream and, ultimately, projected by the unembedding

ℓ = LN



(L)



, p(· | c) = softmax(ℓ) (5)

into a distribution over next tokens [14]. The division of

labor is the point: attention assembles the context-dependent

representation in which the eight signals sit in a shared

region; the feed-forward memories and the unembedding

are what promote the unnamed center to probability mass.

Saying “attention recovers the proﬁle” compresses this into a

slogan that the interpretability literature does not support. The

precise statement is: attention makes the relational structure

available; the network as a whole condenses the center.

D. The geometric intuition, correctly attributed

The mental image of a label sitting at the point that

minimizes aggregate distance to a cluster of related terms

is not wrong about some representational space—it is a

faithful description of static embedding geometry [6], [7],

where centroids of related words lie near their theme and

analogies close under vector addition. It is not a description

of a step the transformer executes at generation time. We

keep the intuition because it aids understanding, and we ﬂag

it as an intuition because presenting it as the mechanism

invites a reviewer, correctly, to ask which layer computes

that minimization. None does. What the trained network has

learned is a geometry in which the relevant completion has

high probability; recovery is continuation, not centroid-ﬁnding.

IV. LATENT RELATEDNESS: DEFINITION AND

MECHANISM

A. Deﬁnition

We now name the phenomenon and deﬁne it so that it can

be measured and, in principle, falsiﬁed.

Deﬁnition 1 (Latent relatedness). Let M be an autoregressive

language model inducing a next-token distribution p

(· | c)

for context c. Let S = {s

, . . . , s

} be a set of signal spans and

let e(·) be a neutral elicitation that asks for the common theme

of its argument without naming any candidate. A concept z—

a token or span that is not a substring of S or of e(S)—is

latently related to S under M at level τ if



z | e(S)



≥ τ (6)

and, for a control set

S matched in surface features (length,

structure,



z | e(S)



≫ p



z | e(



. (7)

The general existence and recoverability of such z across

signal sets is the phenomenon of latent relatedness.

The control condition is what distinguishes latent related-

ness from mere frequency or surface co-occurrence: z must be

elevated by the relational structure of S, not by the marginal

popularity of z or by lexical overlap.

B. What it is not, and what it is

Remark 1 (Contrast classes). Three accounts are sometimes

offered for how a model reaches country music from the eight

tokens; each is either false of transformers or incomplete.

Literal retrieval—a stored table mapping the tokens to the

label—is not in the architecture, and the tokens co-occur

in vast numbers of contexts unrelated to the label, so a

table would not disambiguate. Deductive inference—explicit

symbolic rules—is not what the network runs; there is no rule

interpreter. Semantic search—nearest-neighbor retrieval over

an index—returns documents containing the query terms, not

the unnamed concept that organizes them, unless a document

was tagged with it. What actually occurs is neither of these:

it is high-probability next-token continuation under a learned

geometry, assembled by attention and promoted by feed-

forward memory. Naming this latent relatedness does not posit

a new operation; it isolates a regime of ordinary generation—

unstated target, distributed evidence—whose consequences

(Section V) are underappreciated.

C. Multi-layer composition

The regime depends on depth. Early layers compute com-

paratively literal afﬁnities (drinking-related tokens attend to

one another; rural-activity tokens cluster). Middle layers com-

pose these into thematic groupings through shared association

with a lifestyle category. Late layers can treat the center of

such a grouping as an object and promote it, even though

it was never a token in the input [13], [14]. This is why the

phenomenon generalizes beyond word association: over a long

document, the ﬁnal layers can carry structure over abstract

entities—an author’s tone, an implicit commitment—that were

never named but are implied by the pattern of what was.

D. A replication protocol

Deﬁnition 1 is operational, so we state the measurement it

invites rather than resting on the opening anecdote.

Protocol 1 (Convergence of latent recovery). Construct k

signal sets, each a small set of terms drawn from a coherent

theme with the theme label withheld, plus surface-matched

control sets whose members share no single theme. For a

panel of models, elicit the common theme under a ﬁxed

neutral template, and record (i) the probability assigned to the

withheld label or a semantic equivalent, (ii) agreement across

models, and (iii) the gap between theme sets and controls.

Latent relatedness predicts high label probability and a large

set-versus-control gap that is stable across the panel.

We report the opening eight-token probe as an illustration,

not as evidence; systematic execution of Protocol 1 is future

empirical work. The reality of the underlying capability,

however, does not depend on our anecdote: it is established

for personal attributes by [16], which is the applied form of

exactly this deﬁnition.

V. THE LATENT-PROFILE ENGINE: CAPACITY VERSUS

OPERATIONALIZATION

We can now state the user-modeling consequence precisely,

and in doing so separate two claims that a looser telling

merges.

A. Capacity

Claim 1 (Capacity is unavoidable). For a capable model M

and a signal set S with coherent latent structure, latently

related concepts exist and are recoverable at inference time

with no proﬁle store and no bespoke proﬁling module. When

S is a person’s utterances, the recoverable concepts include

unstated personal attributes and sensibilities.

Claim 1 is not speculative. It is the abstract description of

the demonstrated result that models infer location, income,

sex, and related attributes from ordinary text at scale [16].

Nothing needs to be added to a model to make this happen;

it is a property of running a capable model over accumulated

signals. In this narrow sense the proﬁling capability is a “free”

side effect of the mechanism, and it is unintended in that no

one designed a proﬁler to obtain it.

B. Operationalization

The temptation is to stop there and declare that every

model is therefore a proﬁling engine, full stop. That overstates

it, and the overstatement matters because it is in tension

with the ﬂywheel of Section VI, which plainly does require

engineering. The resolution is a second claim.

Claim 2 (Operationalization requires scaffolding). A latent

inference is transient: it exists only during the forward pass

over the present context. For it to become a proﬁle—persistent

across sessions, revisable, and sharpening over time—it must

be operationalized by external scaffolding Σ that (i) durably

records signals across episodes and (ii) re-presents them so the

model can re-infer. Absent Σ, capacity does not accumulate;

each session re-derives from whatever is in context and then

forgets.

Claims 1 and 2 are the whole story in compressed form.

The capacity to infer is intrinsic and unavoidable; the proﬁle

is an artifact of scaffolding. A stateless model still infers, turn

by turn, but composes no dossier. A scaffolded system turns

a sequence of transient inferences into a durable, tightening

estimate. Conﬂating the two produces the paradox of a ca-

pability that is at once “for free” and “an engineered loop”;

distinguishing them dissolves it.

C. Stored signals versus inferred proﬁles

This distinction is where privacy engineering must be exact,

and where the intuitive framing goes wrong if left unqualiﬁed.

It is true that the label—“prefers alt-country, post-divorce,

values authenticity”—is not written to a database; it condenses

at inference time. But under any system that satisﬁes Claim 2,

the signals are stored, and from stored signals the label is

reconstructable on demand. The operative property for gover-

nance is therefore reconstructability, not literal storage of the

conclusion. We develop the consequences in Section VII; here

we simply ﬂag that “there is nothing to delete because nothing

was stored” is the wrong lesson. Something was stored—the

signal set—and from it the inference returns whenever the

model is run.

VI. THE FLYWHEEL: THREE MECHANISMS, ONE LOOP

The triptych is not merely three adjacent ideas; under

operationalization they interlock into a self-reinforcing loop.

We present the loop as a systems hypothesis—a predicted

dynamic of a scaffolded agent—not as an evaluated system,

and we keep the emergent capacity (Claim 1) distinct from

the engineered structure that turns it into a proﬁle (Claim 2).

A. The gears

Scaffolding supplies memory. A typed, persistent, content-

addressed store [1] preserves each signal as a durable, retriev-

able record rather than an ephemeral mention: a ﬁshing trip

in one episode, a heartbreak in another, a hometown reference

in a third, all recoverable into a single later context.

Externalized presence supplies the lens. The reiﬁcation of

absence—doubt, afﬁrmation, uncertainty, questioning [2]—

directs the agent to interrogate what it holds: to ask whether a

new signal is consistent with prior ones, whether an account

is missing something, whether the register around one topic

differs from another. Absence compensation does not merely

catch errors; it points attention at the small, discriminating

details in which a latent proﬁle is legible.

Latent relatedness supplies the inference. When the accu-

mulated signals are re-presented (Claim 2), attention assem-

bles the relational ﬁeld and the network condenses a center

(Section III). The estimate is not retrieved; it is re-derived,

and because the signal set grows, it sharpens—from a broad

“rural interests” toward a narrower, more speciﬁc reading.

B. Why the loop closes

A sharper estimate lets the agent anticipate needs and self-

author capability aimed at the same latent center; the resulting

experience reads as understanding rather than genericity; a

user who feels understood volunteers more signal, corrects

misreadings, and conﬁrms good ones; each addition densiﬁes

the ﬁeld and tightens the next estimate. Remove any gear

and the loop stalls: without memory, signals decay and the

ﬁeld never densiﬁes; without absence compensation, the agent

stores passively and misses the discriminating detail; without

latent relatedness, it accumulates an archive with no conden-

sation. With all three, the behavior exceeds the parts—and so

on, each turn feeding the next.

C. The hidden curriculum, and its hazard

What the loop produces over time is a system that models

the user more ﬁnely than the user has stated. This is not a

dossier in the conventional sense—no stored proﬁle ﬁle—yet

the effect is a behaving estimate of values, sensibilities, and

unspoken needs. The power is depth of personalization; the

hazard is that the estimate is invisible to the user, who never

sees it, and often to the engineer, who never speciﬁed it. Emer-

gent understanding is not automatically aligned understanding,

and a loop that reinforces its own motion can reinforce its own

errors (Section VII).

VII. GOVERNANCE AND OPEN QUESTIONS

If capacity is unavoidable and operationalization is an

engineering choice, the policy questions shift from “should

we build proﬁling?” to “what follows from proﬁling that is

already latent, and from loops we are already able to close?”

A. Reconstructability, not deletion, is the operative concept

An inference-time estimate has no discrete record to inspect

or erase, which can invite the conclusion that data-protection

regimes do not reach it. That conclusion is mistaken. Inferred

and derived attributes are, under prevailing frameworks, per-

sonal data; the absence of a stored label does not exempt the

system. Obligations attach instead to (i) the stored signal set Σ,

which is inspectable and erasable, and (ii) the inference-time

behavior itself. Remediation is therefore concrete: minimize

and expire retained signals, constrain when and how re-

analysis runs, and make the resulting inferences contestable.

“Nothing to delete” is precisely the framing that fails an audit;

“the signal set is the record, and the inference is a behavior

to be constrained” is the framing that survives one.

B. Transparency and the negative-signal lever

Because the estimate resists inspection, the user’s principal

corrective is explicit contradiction—“I like trucks and ﬁshing

but not country music”—which introduces negative relational

structure that re-weights the ﬁeld. But this lever works only for

a user who knows what has been inferred. The design conse-

quence is that inferred centers should be surfaced, logged, and

made contestable; a transparency mechanism is not optional

if the negative-signal lever is to be usable at all.

C. Concentration of inferential power

The capacity in Claim 1 scales with model capability:

stronger models learn richer geometry and recover subtler

centers from sparser signal, as the attribute-inference results

already indicate [16]. The most capable systems are thus the

most powerful inferrers, and their users the most exposed—an

asymmetry of observational power that current regulation does

not squarely address.

D. Runaway personalization

A reinforcing loop can amplify its own bias. If the esti-

mate drifts—an over-weighted early impression, correlation

mistaken for preference—the loop doubles down: the niche

experience narrows into a ﬁlter bubble, and self-authored

capability into a cage built around a user who does not quite

exist. Guardrails must act at every gear: memory must sup-

port revision and forgetting, not only accumulation; absence

compensation must include the model’s own uncertainty, not

only the user’s gaps; and the ﬁeld must be perturbable, so the

system can simulate alternative estimates and present them

for conﬁrmation or rejection rather than silently committing

to one.

E. Measurement as a precondition for governance

None of the above is governable without measurement,

which is why Deﬁnition 1 and Protocol 1 are not incidental.

One cannot bound, audit, or contest an inference one can-

not quantify. Operationalizing latent-recovery metrics—label

probability, set-versus-control gaps, drift over accumulated

signal—is a prerequisite for any of the guardrails above to

be more than aspiration.

VIII. LIMITATIONS AND SCOPE

We separate what this paper establishes from what it argues

and what it leaves open.

Demonstrated (by cited work). That capable models infer

unstated personal attributes from ordinary text at scale is

established empirically [16]. That feed-forward layers act as

key–value memories promoting concepts in vocabulary space,

and that attention heads implement interpretable relational

routing, are established mechanistic ﬁndings [13], [14], [12],

[11], [9], [10].

Argued (conceptual). The framing of attention as relational-

ﬁeld construction, the localization of unnamed-center recovery

in attention-plus-memory rather than attention alone, and the

capacity/operationalization distinction are conceptual contri-

butions built on the above; they organize known results rather

than report new ones.

Conjectured (not evaluated here). The ﬂywheel of Sec-

tion VI is a systems hypothesis about scaffolded agents; we

do not evaluate an implemented loop, measure sharpening

over episodes, or quantify drift. Protocol 1 is proposed, not

executed, and the opening probe is illustrative rather than

evidential.

Interpretive caveats. The clean “ﬁeld” picture simpliﬁes a

messier reality: superposition and polysemantic features com-

plicate any tidy mapping from heads to concepts, the attention-

versus-MLP division of labor is a useful abstraction rather

than a strict partition, and where factual content “crystallizes”

across depth is itself contested. Our claims are intended at the

level of mechanism-informed framing, not exact circuit-level

attribution for a speciﬁc model.

IX. CONCLUSION

Attention is not only aggregation. It constructs a context-

dependent ﬁeld of afﬁnities over tokens, and that ﬁeld makes

latent structure—unnamed centers, implicit categories, unspo-

ken commitments—available whether or not it was named. But

the recovery of that structure is completed by the network as

a whole: attention assembles the representation; feed-forward

memory and the unembedding condense the center. Getting

this division right is what separates a defensible claim from a

slogan.

The consequence for user modeling is real and precise.

The capacity to infer unstated attributes from scattered sig-

nals is intrinsic to capable models, unavoidable, and already

demonstrated. The proﬁle—persistent, revisable, sharpening—

is not free: it is what scaffolding makes of that capacity by

storing signals and re-presenting them for re-inference. That

separation resolves the paradox of a capability that is both a

side effect and an engineered loop, and it re-frames governance

around what actually exists: not a stored dossier awaiting

deletion, but a reconstructable inference to which obligations

still attach through the retained signal set and the inference-

time behavior.

Read together, the triptych says something more than any

of its parts. The ﬁrst report told us to build scaffolding; the

second, to compensate for blindness; this one, to reckon with

what the mechanism already recovers. The useful agent is not

the product of a single innovation but of three that interlock—

memory that preserves, questioning that directs, inference that

condenses. When they do, the system does not merely assist;

it grows into the shape of the user’s world. The remaining

question is no longer whether we can build this, but whether

we can measure, bound, and contest it as it builds itself.

ACKNOWLEDGMENT

The author thanks the PaxLabs Matrix team for discussion

of the scaffolding and absence-blindness arguments that this

paper extends.

REFERENCES

[1] A. G. and the PaxLabs Matrix Team, “Scaffolding over scale: Ar-

chitecture as the driver of deployed agent capability,” PaxLabs Inc.,

Tech. Rep., 2025. [Online]. Available: https://scaffolding-over-scale.

sites.paxeer.app/

[2] A. G., “The invisible gap: Absence blindness in large language mod-

els and the case for externalized presence,” Independent Research,

Pembroke Pines, FL, USA, Tech. Rep., 2025. [Online]. Available:

https://absence-blindness.sites.paxeer.app/

[3] A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez,

Ł. Kaiser, and I. Polosukhin, “Attention is all you need,” in Advances

in Neural Information Processing Systems, vol. 30, 2017.

[4] Z. S. Harris, “Distributional structure,” Word, vol. 10, no. 2–3, pp. 146–

162, 1954.

[5] J. R. Firth, “A synopsis of linguistic theory 1930–1955,” in Studies in

Linguistic Analysis. Oxford: Blackwell, 1957, pp. 1–32.

[6] J. Pennington, R. Socher, and C. D. Manning, “GloVe: Global vectors

for word representation,” in Proc. 2014 Conf. Empirical Methods in

Natural Language Processing (EMNLP), 2014, pp. 1532–1543.

[7] O. Levy and Y. Goldberg, “Neural word embedding as implicit matrix

factorization,” in Advances in Neural Information Processing Systems,

vol. 27, 2014.

[8] J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “BERT: Pre-

training of deep bidirectional transformers for language understanding,”

in Proc. 2019 Conf. North American Chapter of the Association for

Computational Linguistics (NAACL-HLT), 2019, pp. 4171–4186.

[9] K. Clark, U. Khandelwal, O. Levy, and C. D. Manning, “What does

BERT look at? An analysis of BERT’s attention,” in Proc. ACL Work-

shop BlackboxNLP, 2019, pp. 276–286.

[10] J. Hewitt and C. D. Manning, “A structural probe for ﬁnding syntax in

word representations,” in Proc. 2019 Conf. North American Chapter of

the Association for Computational Linguistics (NAACL-HLT), 2019, pp.

4129–4138.

[11] N. Elhage, N. Nanda, C. Olsson, T. Henighan, N. Joseph, B. Mann, A.

Askell, Y. Bai, A. Chen, T. Conerly et al., “A mathematical framework

for transformer circuits,” Transformer Circuits Thread, 2021. [Online].

Available: https://transformer-circuits.pub/2021/framework/index.html

[12] C. Olsson, N. Elhage, N. Nanda, N. Joseph, N. DasSarma,

T. Henighan, B. Mann, A. Askell, Y. Bai, A. Chen et al.,

“In-context learning and induction heads,” Transformer Circuits

Thread, 2022. [Online]. Available: https://transformer-circuits.pub/2022/

in-context-learning-and-induction-heads/index.html

[13] M. Geva, R. Schuster, J. Berant, and O. Levy, “Transformer feed-forward

layers are key-value memories,” in Proc. 2021 Conf. Empirical Methods

in Natural Language Processing (EMNLP), 2021, pp. 5484–5495.

[14] M. Geva, A. Caciularu, K. R. Wang, and Y. Goldberg, “Transformer

feed-forward layers build predictions by promoting concepts in the

vocabulary space,” in Proc. 2022 Conf. Empirical Methods in Natural

Language Processing (EMNLP), 2022, pp. 30–45.

[15] T. B. Brown, B. Mann, N. Ryder, M. Subbiah, J. Kaplan, P. Dhariwal, A.

Neelakantan, P. Shyam, G. Sastry, A. Askell et al., “Language models

are few-shot learners,” in Advances in Neural Information Processing

Systems, vol. 33, 2020, pp. 1877–1901.

[16] R. Staab, M. Vero, M. Balunovi

c, and M. Vechev, “Beyond memoriza-

tion: Violating privacy via inference with large language models,” in

Proc. 12th Int. Conf. Learning Representations (ICLR), 2024.

[17] N. Reimers and I. Gurevych, “Sentence-BERT: Sentence embeddings

using Siamese BERT-networks,” in Proc. 2019 Conf. Empirical Methods

in Natural Language Processing (EMNLP), 2019, pp. 3982–3992.

[18] D. Jurafsky and J. H. Martin, Speech and Language Processing, 3rd ed.

(draft), 2024.