Back to Embodied Trajectory Systems
working paper2026Anticipation Geometry paper

Anticipation Geometry: Domain-General Trajectory Characterization with Knowledge Graph-Grounded Rewards

Anticipation Geometry defines seven scalar signals over trajectories in arbitrary state spaces. It tests whether motion, conversation, and graph traversal share a measurable geometry of convergence from uncertainty into commitment.

Paper workspace

Live draft structure

working-draft

Artifacts

Markdown paper source

Anticipation Geometry is mapped as source-first until the final paper render is selected.

source-only

Editable source

Paper source exists in Comp-Core and a parallel standalone paper directory. Public PDF render pending.

Source anchors

Comp-Core/papers/anticipation-geometry/paper.md

anticipation-geometry/paper/paper.md

Comp-Core cc-anticipation Rust implementation

Method tags

trajectory geometrytransition pressureKG rewardsdomain-general scalars

Ingest intersections

anticipationtrajectorygeometrykg-rewardtransition-pressure

Status

Paper drafted; Rust implementation exists in Comp-Core; downstream task lift remains the hard proof gate.

Key claims

01

Transition pressure is a reusable signal across domains.

02

Trajectory geometry can become a dense reward layer before final task success is known.

03

The hard remaining proof is downstream task improvement, not only scalar significance.

Public reading note

Working paper public summary safe; attach full text after final review.

Standard skeleton

What this paper must keep proving

Schema

problem

Prediction and confidence are too sparse to monitor reasoning or motion while the trajectory is still unfolding.

method

Define seven scalar signals over vector trajectories and test whether they transfer across motion, conversation, and graph paths.

implementation

Comp-Core cc-anticipation kernel, conversation embeddings, KG path reward experiments, and Rust/Python analysis.

data

Simulated/physical motion, conversation embeddings, and graph-path snapshots with hard negatives.

evaluation

Transition-pressure prediction, KG path discrimination, scalar stability, and downstream reward usefulness.

references

Anticipatory computing, geometric deep learning, CoALA, DSS graph rewards, GraphMERT.

openQuestions

The hard public proof is downstream lift when the geometry is used to improve agent or model decisions.

Checkpoints and references

Proof chain

paperpending

Claim checkpoint

central-claim slot

Every central claim must point to a proof anchor or remain labeled as speculative.

implementationpending

Implementation checkpoint

implementation-map slot

Every method should identify the code path, harness, schema, or protocol that embodies it.

experimentpending

Evidence checkpoint

evidence-manifest slot

Every reported result should point to run IDs, packet IDs, data snapshots, commits, or review artifacts.

external-referencepending

Reference checkpoint

references slot

Every external claim should resolve to a cited paper, benchmark, standard, or documented prior system.

paperpending

Release checkpoint

release-gate slot

Every PDF needs a named condition before it can move from draft to citation-ready.