pith. sign in
Pith Number

pith:DBJPT6B2

pith:2026:DBJPT6B2O6FTK3IXWJHWN56IPU
not attested not anchored not stored refs resolved

Rover: Context-aware Conflict Resolution with LLM

Changhua Luo, Chenxiong Qian, Jiayi Lin, Junzhe Li, Qingyu Zhang

Rover builds multi-layer code graphs so LLMs can resolve merge conflicts more accurately than prior tools or models alone.

arxiv:2605.17279 v1 · 2026-05-17 · cs.SE · cs.AI

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{DBJPT6B2O6FTK3IXWJHWN56IPU}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more

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

Evaluation results show that Rover surpasses all of these approaches in terms of conflict resolution, achieving higher similarity to ground-truth resolutions at character, lexical, and semantic levels.

C2weakest assumption

That the Multi-layer Code Property Graph plus graph connectivity clustering will reliably produce contexts that allow the LLM to infer developer intentions and generate correct resolutions without introducing new errors or hallucinations.

C3one line summary

Rover uses a new Multi-layer Code Property Graph and clustering to supply LLMs with dependency-aware contexts, outperforming standalone LLMs, MergeGen, and WizardMerge on similarity to ground-truth conflict resolutions.

References

50 extracted · 50 resolved · 5 Pith anchors

[1] Paola Accioly, Paulo Borba, and Guilherme Cavalcanti. 2018. Understanding semi-structured merge conflict character- istics in open-source java projects.Empirical Software Engineering23 (2018), 2051–20 2018
[2] gpt-oss-120b & gpt-oss-20b Model Card 2025 · arXiv:2508.10925
[3] Mehdi Ahmed-Nacer, Pascal Urso, and François Charoy. 2013. Improving textual merge result. In9th IEEE International Conference on Collaborative Computing: Networking, Applications and Worksharing. IEE 2013
[5] Waad Aldndni, Na Meng, and Francisco Servant. 2023. Automatic prediction of developers’ resolutions for software merge conflicts.Journal of Systems and Software206 (2023), 111836 2023
[6] Sven Apel, Olaf Leßenich, and Christian Lengauer. 2012. Structured merge with auto-tuning: balancing precision and performance. InProceedings of the 27th IEEE/ACM International Conference on Automated 2012
Receipt and verification
First computed 2026-05-20T00:03:49.465711Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

1852f9f83a778b356d17b24f66f7c87d2885c1697d38e2cce01205f46de995d4

Aliases

arxiv: 2605.17279 · arxiv_version: 2605.17279v1 · doi: 10.48550/arxiv.2605.17279 · pith_short_12: DBJPT6B2O6FT · pith_short_16: DBJPT6B2O6FTK3IX · pith_short_8: DBJPT6B2
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/DBJPT6B2O6FTK3IXWJHWN56IPU \
  | 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: 1852f9f83a778b356d17b24f66f7c87d2885c1697d38e2cce01205f46de995d4
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "f58d77385b95fa9cd072c071d82ee6fb9a971cc184025bc5b32363f4d4df07f2",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.SE",
    "submitted_at": "2026-05-17T06:21:40Z",
    "title_canon_sha256": "4ee047462545419fb480bbfb47f5a7c7b189d0d317a1998c381f9d2595c88a19"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.17279",
    "kind": "arxiv",
    "version": 1
  }
}