Pith Number

pith:NVDKCNLI

pith:2026:NVDKCNLIRHECOSAPLQZK4C4W7H

not attested not anchored not stored refs resolved

Inverse-Hessian Regularization for Continual Learning in ASR

Hugo Van hamme, Steven Vander Eeckt

Inverse-Hessian Regularization adjusts post-fine-tuning ASR updates with prior-task curvature to limit forgetting while preserving adaptability.

arxiv:2601.14751 v1 · 2026-01-21 · eess.AS

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{NVDKCNLIRHECOSAPLQZK4C4W7H}

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

After fine-tuning on a new task, the adaptation is adjusted through a Kronecker-factored inverse Hessian approximation of the previous task, ensuring that the model moves primarily in directions less harmful to past performance, while keeping the method lightweight.

C2weakest assumption

The Kronecker-factored inverse Hessian approximation sufficiently captures the loss landscape curvature of previous ASR tasks to guide safe updates without harming adaptability.

C3one line summary

IHR incorporates curvature information via inverse Hessian approximation into the model merging step for continual learning in ASR, outperforming baselines by reducing forgetting while improving adaptability on two benchmarks.

References

36 extracted · 36 resolved · 1 Pith anchors

[1] To be accurate and inclusive, they must adapt to new speakers, accents, domains, or recording conditions

[2] ASR Model We consider an encoder–decoder ASR model

[3] Inverse-Hessian Regularization for Continual Learning in ASR 2026 · arXiv:2601.14751

[4] More information, including code and detailed results, can be found in our Github repository 1 2048

[5] Experiment 1 As shown by Table 1, our method (IHR) significantly outperforms all baselines, being able to learn with close to zero forgetting (as shown by its -0.1 BWT)

Formal links

1 machine-checked theorem link

Cited by

1 paper in Pith

Inverse-Hessian Regularization for Continual Learning in ASR

Receipt and verification

First computed	2026-05-18T02:45:06.047190Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

6d46a1356889c827480f5c32ae0b96f9ced8148afc032d62441c47eb3183a4a9

Aliases

arxiv: 2601.14751 · arxiv_version: 2601.14751v1 · doi: 10.48550/arxiv.2601.14751 · pith_short_12: NVDKCNLIRHEC · pith_short_16: NVDKCNLIRHECOSAP · pith_short_8: NVDKCNLI

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/NVDKCNLIRHECOSAPLQZK4C4W7H \
  | 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: 6d46a1356889c827480f5c32ae0b96f9ced8148afc032d62441c47eb3183a4a9

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "401d6b3ef01e706b2eab4d5f3817613f2e0e1884527a6b25434053bd24a5671e",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "eess.AS",
    "submitted_at": "2026-01-21T08:10:26Z",
    "title_canon_sha256": "0510dcc5217aebb8613b9b86079c0b49674a4ca1f3ef41dc3b4addab41c07d6a"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2601.14751",
    "kind": "arxiv",
    "version": 1
  }
}