pith. sign in
Pith Number

pith:BI7PFZK3

pith:2026:BI7PFZK3NIRPYKERFDRMPBWRDL
not attested not anchored not stored refs pending

Compute Aligned Training: Optimizing for Test Time Inference

Adam Ousherovitch, Ambuj Tewari

Training LLMs with losses derived from test-time inference operators improves scaling over standard SFT and RL.

arxiv:2604.24957 v2 · 2026-04-27 · cs.LG · cs.AI

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{BI7PFZK3NIRPYKERFDRMPBWRDL}

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

we provide empirical evidence that this training method substantially improves test time scaling over standard training.

C2weakest assumption

That test-time inference strategies can be accurately modeled as operators on the base policy such that the derived losses produce stable and generalizable improvements without introducing new optimization pathologies.

C3one line summary

Compute Aligned Training derives new loss functions by modeling test-time strategies as operators on the base policy, yielding empirical gains in test-time compute scaling over standard SFT and RL.

Cited by

1 paper in Pith

Receipt and verification
First computed 2026-05-21T01:04:26.293762Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

0a3ef2e55b6a22fc289128e2c786d11af363d6746ab2864e2f70a56257fc657f

Aliases

arxiv: 2604.24957 · arxiv_version: 2604.24957v2 · doi: 10.48550/arxiv.2604.24957 · pith_short_12: BI7PFZK3NIRP · pith_short_16: BI7PFZK3NIRPYKER · pith_short_8: BI7PFZK3
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/BI7PFZK3NIRPYKERFDRMPBWRDL \
  | 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: 0a3ef2e55b6a22fc289128e2c786d11af363d6746ab2864e2f70a56257fc657f
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "80192f560d453f9e0541e0741f7046416b90b7693c92be3947d0379fdcebbb68",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-04-27T19:52:38Z",
    "title_canon_sha256": "a6b535bd698bd29fc926fa153ca5747c886bb4c9e9702e9efbb1c3dca6d4af3c"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2604.24957",
    "kind": "arxiv",
    "version": 2
  }
}