pith. sign in
Pith Number

pith:BSYSUPX3

pith:2026:BSYSUPX33HP7ILOERFIPB4YAE6
not attested not anchored not stored refs resolved

GateKD: Confidence-Gated Closed-Loop Distillation for Robust Reasoning

Kasidit Sermsri, Teerapong Panboonyuen

GateKD uses teacher confidence to gate distillation of reasoning steps from large models to smaller ones, creating a closed-loop process that reduces errors.

arxiv:2605.13136 v1 · 2026-05-13 · cs.CL

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{BSYSUPX33HP7ILOERFIPB4YAE6}

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

GateKD consistently outperforms strong open-loop distillation baselines. Notably, GateKD yields substantial gains in logical and symbolic reasoning, remains robust under low-resource distillation settings, and shows clear performance degradation when any gating component is removed.

C2weakest assumption

That teacher confidence scores reliably indicate the correctness of intermediate reasoning steps and that gating on them does not systematically exclude valid but low-confidence reasoning paths or introduce new selection biases.

C3one line summary

GateKD introduces a closed-loop distillation framework that uses teacher confidence to gate soft supervision, hidden-state alignment, and attention transfer, outperforming open-loop baselines on reasoning benchmarks.

References

20 extracted · 20 resolved · 0 Pith anchors

[1] arXiv preprint , volume= 2023
[2] Findings of the Association for Computational Linguistics: EMNLP 2024 , pages= 2024
[3] arXiv preprint , volume= 2023
[4] arXiv preprint , volume= 2024
[5] Proceedings of the 2025 International Natural Language Generation Conference , pages= 2025
Receipt and verification
First computed 2026-05-18T03:08:57.579971Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

0cb12a3efbd9dff42dc48950f0f30027ab36e9a35934acaac6b23cb55135ba80

Aliases

arxiv: 2605.13136 · arxiv_version: 2605.13136v1 · doi: 10.48550/arxiv.2605.13136 · pith_short_12: BSYSUPX33HP7 · pith_short_16: BSYSUPX33HP7ILOE · pith_short_8: BSYSUPX3
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/BSYSUPX33HP7ILOERFIPB4YAE6 \
  | 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: 0cb12a3efbd9dff42dc48950f0f30027ab36e9a35934acaac6b23cb55135ba80
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "b76a13b34e46ab8d52afc292ae5135f6fe6ef417210d29709d7f56ffefdf520b",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by-nc-sa/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2026-05-13T08:04:46Z",
    "title_canon_sha256": "9a72cd8b4fa7239c84a95ca9facc737aeed16753d932afddeef172b7f995639a"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.13136",
    "kind": "arxiv",
    "version": 1
  }
}