# Horizon

The first attempt bent the space with a function nobody trained, and it lost by a wide margin. That was the wrong test of the right question.

Pilot results · d/rksci research · Published 2026-10-08 · Tags: Agent Memory, Retrieval, MemoryAgentBench, LongMemEval, Engram, Hyperbolic Geometry, Embeddings

Does the shape of memory space matter?

Hierarchies grow exponentially and flat space is bad at holding them. Hyperbolic space is built for the job. The question is whether a learned hyperbolic space helps recall for conversation and fact memories, or only for textbook trees.

## 1. Idea

Experiment 6 of 10 in the Engram series. [← Bloodhound](/engram/bloodhound) · [Engram overview](/engram) · [Loom →](/engram/loom)

In flat space, the number of points that fit at a given distance grows polynomially. In a hyperbolic disc it grows exponentially, which is how a tree behaves. Put a tree in the disc with its root at the middle and the leaves near the rim and distances come out right: siblings are close to each other, cousins are far. A memory is not exactly a tree, but “general at the centre, niche at the edge” is a hierarchy.

**Horizon · the disc.** In the Poincaré disc distance stretches towards the rim, leaving room for many niche memories that sit far from each other but near the general one they belong to.

The embedding is trained: it is fitted so that memories that were similar stay close and memories that were general end up near the origin. Retrieval ranks by hyperbolic distance.

## 2. Where it came from

The archived manifold took ordinary embeddings and squashed them into the ball with a tanh function. Nothing was learned. On synthetic clusters, retrieval at five scored 0.348 for hyperbolic distance and 0.296 with “dynamics” added, against 0.876 for cosine. Over twenty cycles of use the hyperbolic version stayed near 0.21 and cosine near 0.95. The project’s own note says it plainly: a naive projection hurts retrieval badly, and cosine is the right baseline until there is a hyperbolic embedding model.

So the result was negative, and it said nothing about hyperbolic space itself: an untrained projection is a strawman. The later variants (a PCA reduction to eight dimensions, an encoder trained by write-read-verify) left no result files, and the project page that advertised “−74% to +4%” has no log behind either number. The synthetic data was also built so that cluster membership was the ground truth, which makes cosine close to an oracle.

## 3. Hypothesis

Stated before running. A Poincaré embedding trained on the similarity graph of each memory does not beat Dense at equal k on these benchmarks, but norm (distance from the origin) tracks how general a memory is, and a small generality prior on top of hyperbolic distance helps on LongMemEval questions that ask for the gist of a topic.

The first half is a prediction that Horizon will not win, made so that the result cannot be spun. If it does win, the prediction will have been wrong in a useful way.

## 4. Design

**Embedding.** A Poincaré embedding in the style of Nickel and Kiela, fitted by Riemannian gradient descent to the similarity graph of one memory’s items (nearest neighbours under bge-small). It is trained per memory, from the memory’s own contents and no labels.

**Retrieval.** Rank by hyperbolic distance from the embedded query. A norm-based generality term is added only in the second analysis, and its weight is fitted on a split disjoint from the pilot sets.

**Equal k.** The same k as every method.

**Seeds.** Training is stochastic: five seeds, mean and range.

**Ablation.** The archived tanh projection, run on the same real questions at last. It shows what the untrained version scores on the data that matters.

## 5. How it is tested

Tier 0 on LongMemEval_S and tier 1 on FactConsolidation, against Dense at equal k. The one figure specific to Horizon is a plot of embedding norm against a measured generality (how many other items an item is a near neighbour of), which shows whether the geometry puts general things in the middle at all, separate from whether it helps retrieval.

The standing rules for every Engram experiment apply here unchanged.

## 6. Results

## 7. What would change the conclusion

- If the trained embedding scores below Dense and the tanh version scores below that, the geometry did not help anywhere, and the archive’s own April note was right.
- If norm does not track generality, the picture in the middle of this page is not what the embedding learned, whatever it scores.
- If the trained embedding matches Dense and the norm tracks generality, the geometry is real and not needed for these benchmarks. That is a finding, not a failure.

## Sources

Experiment 6 of 10 in the Engram series. [← Bloodhound](/engram/bloodhound) · [Engram overview](/engram) · [Loom →](/engram/loom)

Horizon is one of ten experiments in Engram. Its rows come from one run on the pilot sets.