Pith. sign in
Pith Number

pith:AGTVPSHA

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

Can David Beat Goliath? On Multi-Hop Reasoning with Resource-Constrained Agents

Heeyun Jung, Hojae Han, Jongyoon Kim, Seung-won Hwang

David-GRPO lets small agents improve multi-hop QA by injecting expert trajectories and scoring evidence coverage during low-batch RL.

arxiv:2601.21699 v3 · 2026-01-29 · cs.CL · cs.IR

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

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

On agents up to 1.5B parameters trained on four RTX 3090 GPUs, David-GRPO improves over prior RL baselines under the same low-budget setting on six multi-hop QA benchmarks.

C2weakest assumption

That a small number of injected expert trajectories plus evidence-coverage scoring will reliably increase retrieval depth and evidence coverage without introducing bias or overfitting in the low-batch regime.

C3one line summary

David-GRPO improves low-budget RL training for multi-hop QA agents by bootstrapping expert trajectories and converting on-policy partial successes into evidence-coverage signals that increase retrieval depth.

References

12 extracted · 12 resolved · 2 Pith anchors

[1] URL https://doi.org/10.1145/ 2959100.2959190 2024 · doi:10.1145/2959100.2959190
[2] The Llama 3 Herd of Models 2020 · doi:10.18653/v1/2020.emnlp-main.710
[3] Jacovi, A., Caciularu, A., Goldman, O., and Goldberg, Y 2020 · doi:10.18653/v1/2023.emnlp-main
[4] emnlp-main.495/ 2023 · doi:10.1145/3701716.3715313
[5] Qwen2.5 Technical Report 2025 · doi:10.18653/v1/2025
Receipt and verification
First computed 2026-08-20T01:22:07.080492Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

01a757c8e099497b3192000008b90370263490149ca715198c02dbdbf91b9744

Aliases

arxiv: 2601.21699 · arxiv_version: 2601.21699v3 · doi: 10.48550/arxiv.2601.21699 · pith_short_12: AGTVPSHATFEX · pith_short_16: AGTVPSHATFEXWMMS · pith_short_8: AGTVPSHA
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/AGTVPSHATFEXWMMSAAAAROIDOA \
  | 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: 01a757c8e099497b3192000008b90370263490149ca715198c02dbdbf91b9744
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "836dd711ba9b9e3de66a380dee989d64b6bd58ca715ac2a3fb025e3fb12c636d",
    "cross_cats_sorted": [
      "cs.IR"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2026-01-29T13:31:28Z",
    "title_canon_sha256": "9140bcf4b9d78d62499ba28c909f731da070284c3f9b3bd9184b589cf7e37981"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2601.21699",
    "kind": "arxiv",
    "version": 3
  }
}