Pith Number

pith:YB3NHDCI

pith:2026:YB3NHDCIAU4LOID55LM4DUZCQ4

not attested not anchored not stored refs resolved

When to Think Fast and Slow? AMOR: Adaptive Entropy Gate for Hybrid Models

Chen Shani, Haoran Zheng

AMOR routes attention to high-entropy tokens in recurrent models, matching full hybrids while using attention on only 22 percent of positions.

arxiv:2602.13215 v2 · 2026-01-22 · cs.AI

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{YB3NHDCIAU4LOID55LM4DUZCQ4}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

Across Mamba2 and Gated DeltaNet backbones (180M-1.5B), AMOR consistently matches or outperforms both pure recurrent models and fixed-schedule hybrid baselines while invoking attention on only ~22% of tokens.

C2weakest assumption

That a dynamic threshold derived from running batch median and scaled standard deviation of output entropy reliably identifies positions where recurrent state alone is insufficient, without introducing distribution-shift artifacts or requiring per-task retuning.

C3one line summary

AMOR uses output entropy to gate attention in recurrent hybrids, matching full attention performance at roughly 22% attention invocations across 180M-1.5B models.

References

17 extracted · 17 resolved · 11 Pith anchors

[1] Pondernet: Learning to ponder

[2] Longformer: The Long-Document Transformer 2004 · arXiv:2004.05150

[3] The Consciousness Prior, Dec

[4] Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation · arXiv:1308.3432

[5] Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality · arXiv:2405.21060

Formal links

2 machine-checked theorem links

Receipt and verification

First computed	2026-05-18T03:09:23.557862Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

c076d38c480538b7207dead9c1d322872d8718073469adce944c2c128e623925

Aliases

arxiv: 2602.13215 · arxiv_version: 2602.13215v2 · doi: 10.48550/arxiv.2602.13215 · pith_short_12: YB3NHDCIAU4L · pith_short_16: YB3NHDCIAU4LOID5 · pith_short_8: YB3NHDCI

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/YB3NHDCIAU4LOID55LM4DUZCQ4 \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: c076d38c480538b7207dead9c1d322872d8718073469adce944c2c128e623925

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "0676e5e3bb78eef29abe6f00ccc137b3f922cecc1e1654925b39f7e25c252d9e",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.AI",
    "submitted_at": "2026-01-22T17:19:58Z",
    "title_canon_sha256": "e0f05bdcd9b0fe30d0eceb440cd666cdfc4f4d5bdb61b11b069921a5723ec0c0"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2602.13215",
    "kind": "arxiv",
    "version": 2
  }
}