Skip to content
← Back to explorer

FeatBench: Towards More Realistic Evaluation of Feature-level Code Generation

Haorui Chen, Chengze Li, Jia Li · Sep 26, 2025 · Citations: 0

Open arXiv Find Implementation RSS feed Shortlist (0)

Data freshness

Extraction: Fresh

Check recency before relying on this page for active eval decisions. Use stale pages as context and verify against current hub results.

Metadata refreshed

Feb 18, 2026, 3:49 PM

Stale

Extraction refreshed

Apr 13, 2026, 6:45 AM

Fresh

Extraction source

Persisted extraction

Confidence 0.25

Abstract

Evaluating Large Language Models (LLMs) on repository-level feature implementation is a critical frontier in software engineering. However, establishing a benchmark that faithfully mirrors realistic development scenarios remains a significant challenge. Existing feature-level benchmarks generally suffer from two primary limitations: unrealistic task inputs enriched with code hints and significant data leakage risks due to their static nature. To address these limitations, we propose a new benchmark - FeatBench, which introduces the following advances: (1) Realistic Task Inputs. Task inputs consist solely of natural language requirements, strictly devoid of code hints (e.g., function signatures). This format mirrors realistic software development by requiring agents to independently bridge the gap between abstract user intent and concrete code changes. (2) Evolving Data. FeatBench employs a fully automated pipeline to construct new benchmark versions from the latest repositories, effectively mitigating data contamination. The initial release comprises 157 tasks sourced from 27 actively maintained repositories. We evaluate two state-of-the-art agent frameworks with four leading LLMs on FeatBench. The results reveal that FeatBench poses a significant challenge, with the highest resolved rate reaching only 29.94%. Crucially, our analysis uncovers a prevalent behavioral pattern of aggressive implementation, which leads to "scope creep" and widespread regressions where agents break existing features by diverging from the user's explicit intent. We release FeatBench, our automated pipeline, and all experimental results to facilitate further community research.

Low-signal caution for protocol decisions

Use this page for context, then validate protocol choices against stronger HFEPX references before implementation decisions.

  • Extraction flags indicate low-signal or possible false-positive protocol mapping.
  • Extraction confidence is 0.25 (below strong-reference threshold).
  • No explicit evaluation mode was extracted from available metadata.
Human Eval Hub LLM-as-Judge Hub Pairwise Preference Hub

HFEPX Relevance Assessment

This paper is adjacent to HFEPX scope and is best used for background context, not as a primary protocol reference.

Best use

Background context only

Use if you need

Background context only.

Main weakness

Extraction flags indicate low-signal or possible false-positive protocol mapping.

Trust level

Low

Eval-Fit Score

0/100 • Low

Treat as adjacent context, not a core eval-method reference.

Human Feedback Signal

Not explicit in abstract metadata

Evaluation Signal

Weak / implicit signal

HFEPX Fit

Adjacent candidate

Extraction confidence: Low

If you are doing eval pipeline work, start here:

Human Eval Hub LLM-as-Judge Hub Pairwise Preference Hub Tool-Use Eval Hub

Field Provenance & Confidence

Each key protocol field shows extraction state, confidence band, and data source so you can decide whether to trust it directly or validate from full text.

Human Feedback Types

missing

None explicit

Confidence: Low Source: Persisted extraction missing

No explicit feedback protocol extracted.

Evidence snippet: Evaluating Large Language Models (LLMs) on repository-level feature implementation is a critical frontier in software engineering.

Evaluation Modes

missing

None explicit

Confidence: Low Source: Persisted extraction missing

Validate eval design from full paper text.

Evidence snippet: Evaluating Large Language Models (LLMs) on repository-level feature implementation is a critical frontier in software engineering.

Quality Controls

missing

Not reported

Confidence: Low Source: Persisted extraction missing

No explicit QC controls found.

Evidence snippet: Evaluating Large Language Models (LLMs) on repository-level feature implementation is a critical frontier in software engineering.

Benchmarks / Datasets

partial

Featbench

Confidence: Low Source: Persisted extraction evidenced

Useful for quick benchmark comparison.

Evidence snippet: To address these limitations, we propose a new benchmark - FeatBench, which introduces the following advances: (1) Realistic Task Inputs.

Reported Metrics

missing

Not extracted

Confidence: Low Source: Persisted extraction missing

No metric anchors detected.

Evidence snippet: Evaluating Large Language Models (LLMs) on repository-level feature implementation is a critical frontier in software engineering.

Rater Population

missing

Unknown

Confidence: Low Source: Persisted extraction missing

Rater source not explicitly reported.

Evidence snippet: Evaluating Large Language Models (LLMs) on repository-level feature implementation is a critical frontier in software engineering.

Human Data Lens

  • Uses human feedback: No
  • Feedback types: None
  • Rater population: Unknown
  • Unit of annotation: Unknown
  • Expertise required: Coding
  • Extraction source: Persisted extraction

Evaluation Lens

  • Evaluation modes:
  • Agentic eval: None
  • Quality controls: Not reported
  • Confidence: 0.25
  • Flags: low_signal, possible_false_positive

Protocol And Measurement Signals

Benchmarks / Datasets

Featbench

Reported Metrics

No metric terms were extracted from the available abstract.

Research Brief

Deterministic synthesis

However, establishing a benchmark that faithfully mirrors realistic development scenarios remains a significant challenge. HFEPX protocol signal is limited in abstract-level metadata, so treat it as adjacent context. Updated from current HFEPX corpus.

Generated Apr 13, 2026, 6:45 AM · Grounded in abstract + metadata only

Key Takeaways

  • However, establishing a benchmark that faithfully mirrors realistic development scenarios remains a significant challenge.
  • To address these limitations, we propose a new benchmark - FeatBench, which introduces the following advances: (1) Realistic Task Inputs.

Researcher Actions

  • Treat this as method context, then pivot to protocol-specific HFEPX hubs.
  • Cross-check benchmark overlap: Featbench.
  • Verify metric definitions before comparing against your eval pipeline.

Caveats

  • Generated from title, abstract, and extracted metadata only; full-paper implementation details are not parsed.
  • Low-signal flag detected: protocol relevance may be indirect.

Research Summary

Contribution Summary

  • However, establishing a benchmark that faithfully mirrors realistic development scenarios remains a significant challenge.
  • To address these limitations, we propose a new benchmark - FeatBench, which introduces the following advances: (1) Realistic Task Inputs.
  • We evaluate two state-of-the-art agent frameworks with four leading LLMs on FeatBench.

Why It Matters For Eval

  • To address these limitations, we propose a new benchmark - FeatBench, which introduces the following advances: (1) Realistic Task Inputs.
  • We evaluate two state-of-the-art agent frameworks with four leading LLMs on FeatBench.

Researcher Checklist

  • Gap: Human feedback protocol is explicit

    No explicit human feedback protocol detected.

  • Gap: Evaluation mode is explicit

    No clear evaluation mode extracted.

  • Gap: Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Pass: Benchmark or dataset anchors are present

    Detected: Featbench

  • Gap: Metric reporting is present

    No metric terms extracted.

Category-Adjacent Papers (Broader Context)

These papers are nearby in arXiv category and useful for broader context, but not necessarily protocol-matched to this paper.

Need human evaluators for your AI research? Scale annotation with expert AI Trainers.

Post a Job Get a Quote