pith. sign in
Pith Number

pith:BWIAXPTY

pith:2024:BWIAXPTYFCWBSFU2474RYSMEMR
not attested not anchored not stored refs resolved

Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference

Baris Kasikci, Guangxuan Xiao, Jiaming Tang, Kan Zhu, Song Han, Yilong Zhao

Quest selects only the top-K critical KV cache pages using query vectors and min-max key bounds to accelerate long-context LLM attention.

arxiv:2406.10774 v2 · 2024-06-16 · cs.CL · cs.LG

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{BWIAXPTYFCWBSFU2474RYSMEMR}

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

By only loading the Top-K critical KV cache pages for attention, Quest significantly speeds up self-attention without sacrificing accuracy. We show that Quest can achieve up to 2.23x self-attention speedup, which reduces inference latency by 7.03x while performing well on tasks with long dependencies with negligible accuracy loss.

C2weakest assumption

That the min/max key approximation per page, combined with query-vector scoring, reliably identifies the truly critical pages without dropping information that would change the final attention output on long-dependency tasks.

C3one line summary

Quest speeds up long-context LLM self-attention by up to 2.23x via query-dependent selection of top-K critical KV cache pages, cutting overall latency by 7.03x with negligible accuracy loss.

References

72 extracted · 72 resolved · 0 Pith anchors

[1] I ntroducing the next generation of C laude 2024
[2] Longbench: A bilingual, multitask benchmark for long context understanding, 2023 2023
[3] Y., Ermon, S., Rudra, A., and Ré, C 2022
[4] A., and Gardner, M 2021
[5] Model tells you what to discard: Adaptive kv cache compression for llms, 2024 2024

Cited by

43 papers in Pith

Receipt and verification
First computed 2026-05-17T23:38:52.375607Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

0d900bbe7828ac19169ae7f91c49846449e053ca56781d809c9ea37308fe874f

Aliases

arxiv: 2406.10774 · arxiv_version: 2406.10774v2 · doi: 10.48550/arxiv.2406.10774 · pith_short_12: BWIAXPTYFCWB · pith_short_16: BWIAXPTYFCWBSFU2 · pith_short_8: BWIAXPTY
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/BWIAXPTYFCWBSFU2474RYSMEMR \
  | 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: 0d900bbe7828ac19169ae7f91c49846449e053ca56781d809c9ea37308fe874f
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "f1e057c54ef5bad63ddb13e05dfd763685b60792f4462cce20f8292321c5e5bb",
    "cross_cats_sorted": [
      "cs.LG"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2024-06-16T01:33:02Z",
    "title_canon_sha256": "d3fe8b74b07aadd7ad629906ecb00045739da53c44bcccc04a230ee451b8b27b"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2406.10774",
    "kind": "arxiv",
    "version": 2
  }
}