Pith Number

pith:P3WII2RI

pith:2026:P3WII2RIRDMM2E23I6IOJXPS54

not attested not anchored not stored refs resolved

GQLA: Group-Query Latent Attention for Hardware-Adaptive Large Language Model Decoding

Fanxu Meng

Group-Query Latent Attention exposes two equivalent decoding paths from one set of weights for hardware-specific LLM inference.

arxiv:2605.15250 v1 · 2026-05-14 · cs.LG · 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{P3WII2RIRDMM2E23I6IOJXPS54}

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

A single set of GQLA weights pins the rooflines of both H100 (MQA-absorb, s_q=1) and H20 (GQA + MTP, s_q=2), while supporting up to 8-way zero-redundancy tensor parallelism on the GQA path, and compresses the per-token KV cache to 28.125% of the GQA baseline on the MQA-absorb path.

C2weakest assumption

The two decoding paths remain algebraically equivalent after the TransGQLA conversion from a pretrained GQA checkpoint, so that accuracy and the claimed cache compression are preserved without any retraining or additional fine-tuning steps.

C3one line summary

GQLA exposes two algebraically equivalent decoding paths over one set of weights so a single model can hit roofline on both high-end and commodity GPUs while cutting KV cache size to 28% on the absorbed path.

References

18 extracted · 18 resolved · 7 Pith anchors

[1] Fast Transformer Decoding: One Write-Head is All You Need 1911 · arXiv:1911.02150

[2] Proceedings of EMNLP , year=

[3] DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model · arXiv:2405.04434

[4] Advances in Neural Information Processing Systems , volume=

[5] DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models · arXiv:2512.02556

Receipt and verification

First computed	2026-05-20T00:00:48.500751Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

7eec846a2888d8cd135b4790e4ddf2ef1a7cd54f3e4f2fccfb408dec59d6e21c

Aliases

arxiv: 2605.15250 · arxiv_version: 2605.15250v1 · doi: 10.48550/arxiv.2605.15250 · pith_short_12: P3WII2RIRDMM · pith_short_16: P3WII2RIRDMM2E23 · pith_short_8: P3WII2RI

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/P3WII2RIRDMM2E23I6IOJXPS54 \
  | 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: 7eec846a2888d8cd135b4790e4ddf2ef1a7cd54f3e4f2fccfb408dec59d6e21c

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "95e6ae5ced3e7156f249893b3ea3a5660d173502854936614102cd22d1e7d013",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-05-14T15:50:01Z",
    "title_canon_sha256": "084638641901ed903a190caf883afa608874d335be0d4c7f42c3e49e1b2745dc"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.15250",
    "kind": "arxiv",
    "version": 1
  }
}