Pith Number

pith:SPCBNWHG

pith:2026:SPCBNWHGX7OWVE5HW5YXQVOBGX

not attested not anchored not stored refs resolved

Hydra: Efficient, Correct Code Generation via Checkpoint-and-Rollback Support

Alexander Du, Danyang Zhuo, Jianjun Ou, Matthew Lentz

Hydra enables asynchronous compile checks and targeted rollback repairs for LLM-generated C/C++ code.

arxiv:2605.15238 v1 · 2026-05-14 · cs.SE · cs.AI · cs.PL

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{SPCBNWHGX7OWVE5HW5YXQVOBGX}

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

Paired with a token-efficient repair strategy, Hydra reduces latency by up to 71% and token consumption by up to 70% relative to post-hoc repair on C/C++ code generation tasks that encounter static errors.

C2weakest assumption

That retrofitting the Clang C/C++ compiler with modest modifications is sufficient to provide reliable checkpoint-and-rollback support without introducing substantial overhead or breaking compatibility for typical code generation workloads.

C3one line summary

Hydra enables asynchronous static error checking and targeted checkpoint-rollback repair during LLM code generation, cutting latency by up to 71% and token use by up to 70% versus post-hoc repair on C/C++ tasks.

References

42 extracted · 42 resolved · 3 Pith anchors

[1] Lakshya Agrawal, Aditya Kanade, Navin Goyal, Shuvendu K Lahiri, and Sriram Rajamani. 2023. Monitor-guided decoding of code LMs with static analysis of repository context. InConference on Neural Inform 2023

[2] Anthropic. 2026. Claude Code.https://claude.com/product/claude- codeAccessed: 2026-04-09 2026

[3] InFindings of the Association for Computational Linguistics: EMNLP 2024, Yaser Al-Onaizan, Mohit Bansal, and Yun-Nung Chen (Eds.) 2024 · doi:10.18653/v1/2024.findings-

[4] Statically contextualizing large language models with typed holes 2024 · doi:10.1145/3689728

[5] Test Intention Guided LLM-Based Unit Test Generation 2025

Receipt and verification

First computed	2026-05-20T00:00:47.866234Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

93c416d8e6bfdd6a93a7b7717855c135d7b3eb95dc6a47f843ea5025475be76c

Aliases

arxiv: 2605.15238 · arxiv_version: 2605.15238v1 · doi: 10.48550/arxiv.2605.15238 · pith_short_12: SPCBNWHGX7OW · pith_short_16: SPCBNWHGX7OWVE5H · pith_short_8: SPCBNWHG

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/SPCBNWHGX7OWVE5HW5YXQVOBGX \
  | 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: 93c416d8e6bfdd6a93a7b7717855c135d7b3eb95dc6a47f843ea5025475be76c

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "e6b887b5a2d0d9c05a9a4fb154a132205430ae9beea2bd85e39bfb952ed308a1",
    "cross_cats_sorted": [
      "cs.AI",
      "cs.PL"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.SE",
    "submitted_at": "2026-05-14T03:18:16Z",
    "title_canon_sha256": "89af96dffa207f814395c1be1e965d1aeec62dbae698d1c1bb69135258e2e1ae"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.15238",
    "kind": "arxiv",
    "version": 1
  }
}