Pith. sign in

REVIEW 7 cited by

RustEvo²: An Evolving Benchmark for API Evolution in LLM-based Rust Code Generation

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2503.16922 v1 pith:RIF5AMLS submitted 2025-03-21 cs.SE cs.AI

RustEvo^2: An Evolving Benchmark for API Evolution in LLM-based Rust Code Generation

classification cs.SE cs.AI
keywords changesmodelsrustapisbenchmarksllmsrustevocode
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Large Language Models (LLMs) have become pivotal tools for automating code generation in software development. However, these models face significant challenges in producing version-aware code for rapidly evolving languages like Rust, where frequent Application Programming Interfaces (API) changes across versions lead to compatibility issues and correctness errors. Existing benchmarks lack systematic evaluation of how models navigate API transitions, relying on labor-intensive manual curation and offering limited version-specific insights. To address this gap, we present RustEvo, a novel framework for constructing dynamic benchmarks that evaluate the ability of LLMs to adapt to evolving Rust APIs. RustEvo automates dataset creation by synthesizing 588 API changes (380 from Rust standard libraries, 208 from 15 third-party crates) into programming tasks mirroring real-world challenges. These tasks cover four API evolution categories: Stabilizations, Signature Changes, Behavioral Changes, and Deprecations, reflecting their actual distribution in the Rust ecosystem. Experiments on state-of-the-art (SOTA) LLMs reveal significant performance variations: models achieve a 65.8% average success rate on stabilized APIs but only 38.0% on behavioral changes, highlighting difficulties in detecting semantic shifts without signature alterations. Knowledge cutoff dates strongly influence performance, with models scoring 56.1% on before-cutoff APIs versus 32.5% on after-cutoff tasks. Retrieval-Augmented Generation (RAG) mitigates this gap, improving success rates by 13.5% on average for APIs released after model training. Our findings underscore the necessity of our evolution-aware benchmarks to advance the adaptability of LLMs in fast-paced software ecosystems. The framework and the benchmarks are publicly released at https://github.com/SYSUSELab/RustEvo.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. When Retrieval Hurts Code Completion: A Diagnostic Study of Stale Repository Context

    cs.SE 2026-05 accept novelty 7.0

    Stale repository context in code RAG actively induces models to produce obsolete helper references, raising stale outputs by 76-88 percentage points over current-only retrieval in a 17-sample diagnostic study.

  2. Automating Database-Native Function Code Synthesis with LLMs

    cs.DB 2026-04 conditional novelty 7.0

    DBCooker automates synthesis of database native functions via LLM-guided characterization, coding plans, hybrid filling, and progressive validation, delivering 34.55% higher accuracy than baselines on SQLite, PostgreS...

  3. Diagnosing Knowledge Gaps in LLM Tool Use: An Agentic Benchmark for Novel API Acquisition

    cs.AI 2026-06 unverdicted novelty 6.0

    NovelAPIBench is a dynamic benchmark that decomposes novel-API knowledge into bundles, generates executable tasks, and shows usage examples as the strongest signal while retrieval and parametric adaptation play comple...

  4. Bridging Generation and Training: A Systematic Review of Quality Issues in LLMs for Code

    cs.SE 2026-05 accept novelty 6.0

    A review of 114 studies creates taxonomies for code and data quality issues, formalizes 18 propagation mechanisms from training data defects to LLM-generated code defects, and synthesizes detection and mitigation techniques.

  5. When LLMs Lag Behind: Knowledge Conflicts from Evolving APIs in Code Generation

    cs.SE 2026-04 unverdicted novelty 6.0

    LLMs produce executable code only 42.55% of the time under API evolution without full documentation, improving to 66.36% with structured docs and by 11% more with reasoning strategies, yet outdated patterns persist.

  6. Qiskit Code Migration with LLMs

    cs.SE 2026-06 unverdicted novelty 5.0

    A taxonomy-guided RAG system with LLMs reduces hallucinations and improves migration suggestions for Qiskit code compared to unconstrained retrieval.

  7. Meta-Tool: Efficient Few-Shot Tool Adaptation for Small Language Models

    cs.CL 2026-04 unverdicted novelty 4.0

    A 3B model with few-shot prompting reaches 79.7% of GPT-5 tool-use performance while a hypernetwork adaptation adds zero measurable benefit across four benchmarks.