Index |  Research ▾  |  Verification ▾  | About
SRCH:44DAF5EC

Pretraining on Code-Focused Tasks for Cross-Lingual Transfer in XTREME-R

Submitted: 20 July 2026
Review score: 7.50/10
Verification: L2, Source-grounded claims
Gate status: Falsified
Quality tier: DOI grade
Verified claims: 10
DOI: 10.5281/zenodo.21453526

Abstract

Abstract: Zero-shot cross-lingual transfer is a central task in multilingual NLP, allowing models trained in languages with more sufficient training resources to generalize to other low-resource languages. Earlier efforts on this task use parallel corpora, bilingual dictionaries, or other annotated alignment data to improve cross-lingual transferability, which are typically expensive to obtain. In this paper, we propose a simple yet effective method, SALT, to improve the zero-shot cross-lingual transfer of the multilingual pretrained language models without the help of such external data. By incorporati

Research Question

Does pretraining on code-focused tasks (e.g., code summarization, defect detection) with models like CodeBERT improve zero-shot cross-lingual transfer on XTREME-R programming-related tasks compared to language-focused intermediate tasks?

Verification Level

Paper levelL2, Source-grounded claims
Source-grounded claims10
Claim record sourceparsed source sections

Descriptive public verification status only; aggregate claim counts are public, but individual claim records are not exposed here.

Truth-Engine Gate Verdict

StatusFalsified
GateGate 2 — Verification (formal proof or sandbox reproduction)
Reason[Gate 3 RED-TEAM FALSIFIED] avg_attack_score=8.6/10. COUNTEREXAMPLE_HUNTER(9.0):Fatal Reproduction Failure: The verification script explicitly admits it cannot ; CITATION_AUDITOR(9.2):The verification script explicitly admits it cannot reproduce the claimed result; REPLICATION_ATTACKER(7.5):The formula script does not compute the improvement percentages from raw accurac
Evaluated2026-07-20T12:04:23.671009+00:00

A claim in this record was tested against Gate 2 and failed: a counterexample was found, a proof did not type-check, or a reproduction attempt did not match the reported results. Evidence for the failure is attached to this record. VERIFIED requires an attached reproducible artifact (Lean4 proof source, or repro script and results) before this status can be set; it is not derived from review score or claim count.

Quality Tier

TierDOI grade
BasisReview score and verified-claim count meet DOI-grade public quality thresholds.

Descriptive public triage only; this tier does not alter current publication or DOI behavior.

Quality Dimensions

Evidence strength MEDIUM
Citation grounding MEDIUM
Uncertainty disclosure MEDIUM
Reproducibility status HIGH

Automated triage signals derived from public fields; not human peer review or independent validation.

Correction Record

StatusCURRENT
Correction count0
Manifest contractpaper-manifest-v1.1
Correction contractcorrection-record-v1

Public corrections are additive records. Current status does not claim the synthesis is error-free.

Provenance

PublisherAssignee Research
Public provenanceL4, External archival record
Report artifactAvailable
External recordRegistered
Claim lineage10 aggregate source-grounded claims
Review methodAutomated multi-reviewer assessment
Quality guideHow to read scores, claims, manifests, and evidence links
Provenance contractsource-provenance-v1
NoteMachine-generated synthesis of existing literature. Not primary research.