Skip to content
OpenTrain AIFor AI Companies

HFEPX · Eval paper review

Peer Pressure: Execution-Verified Co-Evolution of Attacker and Defender Populations for Code Security

Khushbu Pahwa, Yubin Ge, Daniele Bonadiman, Xiaofei Ma +1 more

Published

Oct 17, 2026

Citations

0

Trust level

Moderate

Usefulness score

65/100 (Medium)

Extraction confidence

70% (Moderate)

Derived from extracted protocol signals and abstract evidence.

Rater population

Not reported

Signals refreshed

Oct 17, 2026

Open arXiv Find implementation Shortlist (0) RSS feed

Tags

Automatic Metrics Coding Red Team

Should you rely on this paper?

This paper has useful evaluation signal, but protocol completeness is partial; pair it with related papers before deciding implementation strategy.

Use this for comparison and orientation, not as your only source.

Best use

Secondary protocol comparison source

Use if you need

A secondary eval reference to pair with stronger protocol papers.

What to verify

Validate the evaluation procedure and quality controls in the full paper before operational use.

Main weakness

No major weakness surfaced.

Human feedback signal
Detected
From extracted signals
Evaluation signal
Detected
Eval setup described
Usefulness for eval research
65/100
Moderate-confidence candidate

Useful as a secondary reference; validate protocol details against neighboring papers.

If you are doing eval pipeline work, start here

Human Eval Papers LLM-as-Judge Papers Pairwise Preference Papers Tool-Use Evaluation Search

Abstract

LLM code agents increasingly ship code with little oversight, yet defenders trained on static vulnerability datasets fall behind adapting adversaries. We co-train attacker and defender instead, and let execution decide every reward: an attacker (Red) edits a real repository and earns nothing unless its patch applies, builds, passes the tests, completes its development task and reproduces a real exploit in an isolated container, whose path also labels the entry point, the critical operation and the trace between them. Peer-Relative Policy Optimization (PRPO) makes both sides populations of LoRA adapters over one frozen base and credits each against same-role peers that faced an identical condition, placing the population inside the policy-gradient baseline rather than the opponent pool, with a tournament preserving diversity. Matching cancels the condition's difficulty: at equal coverage it is worth 2.3 points of balanced accuracy over a baseline computed on other conditions, and it matches a within-policy control at half the verifier spend. PRPO's 7B diff reviewer reaches 64.8% balanced accuracy on four unseen applications in three untrained languages, 5.5 points above supervised security fine-tuning and 4.4 above a frozen frontier reviewer (Claude Opus 4.8), which it leads by 15.8 on the application it trains on. Its 9B repository agent recovers 23 of 172 held-out VulnGym advisories, against 10 after supervised fine-tuning and 7 for the same backbone untrained under the same scaffold, and locates 17 of 43 reachable entry points against 13 for mini-swe-agent with GLM-5.2. Run unchanged on jailbreak defense, the same algorithm matches or exceeds WildGuard's static F1 at equal corpus and model size. Under a judge distinct from its training oracle it cuts attack success from 36.0% to 4.6%, against WildGuard's 8.9%.

What we could verify

These are the protocol signals we could actually recover from the available paper metadata. Use them to decide whether this paper is worth deeper reading.

Human Feedback Types

strong

Red Team

Directly usable for protocol triage.

"LLM code agents increasingly ship code with little oversight, yet defenders trained on static vulnerability datasets fall behind adapting adversaries."

Evaluation Modes

strong

Automatic Metrics

Includes extracted eval setup.

"LLM code agents increasingly ship code with little oversight, yet defenders trained on static vulnerability datasets fall behind adapting adversaries."

Quality Controls

missing

Not reported

No explicit QC controls found.

"LLM code agents increasingly ship code with little oversight, yet defenders trained on static vulnerability datasets fall behind adapting adversaries."

Benchmarks / Datasets

missing

Not extracted

No benchmark anchors detected.

"LLM code agents increasingly ship code with little oversight, yet defenders trained on static vulnerability datasets fall behind adapting adversaries."

Reported Metrics

strong

Accuracy, F1

Useful for evaluation criteria comparison.

"Matching cancels the condition's difficulty: at equal coverage it is worth 2.3 points of balanced accuracy over a baseline computed on other conditions, and it matches a within-policy control at half the verifier spend."

Benchmarks and datasets

No benchmark or dataset names were extracted from the available abstract.

Reported metrics

accuracyf1
Human feedback details
Uses human feedback
Yes
Feedback types
Red Team
Rater population
Not reported
Expertise required
Coding
Evaluation details
Evaluation modes
Automatic Metrics
Agentic eval
None
Quality controls
Not reported
Evidence quality
Moderate
Use this page as
Secondary protocol comparison source

Research brief

Metadata summary

LLM code agents increasingly ship code with little oversight, yet defenders trained on static vulnerability datasets fall behind adapting adversaries.

Based on abstract + metadata only. Check the source paper before making high-confidence protocol decisions.

Key takeaways

  • LLM code agents increasingly ship code with little oversight, yet defenders trained on static vulnerability datasets fall behind adapting adversaries.
  • We co-train attacker and defender instead, and let execution decide every reward: an attacker (Red) edits a real repository and earns nothing unless its patch applies, builds, passes the tests, completes its development task and reproduces a real exploit in an isolated container, whose path also labels the entry point, the critical operation and the trace between them.
  • Peer-Relative Policy Optimization (PRPO) makes both sides populations of LoRA adapters over one frozen base and credits each against same-role peers that faced an identical condition, placing the population inside the policy-gradient baseline rather than the opponent pool, with a tournament preserving diversity.

Researcher actions

  • Compare this paper against nearby papers in the same arXiv category before using it for protocol decisions.
  • Validate inferred eval signals (Automatic metrics) against the full paper.
  • Use related-paper links to find stronger protocol-specific references.

Caveats

  • Generated from abstract + metadata only; no PDF parsing.
  • Signals below are heuristic and may miss details reported outside the abstract.

Recommended queries

Accuracy

Contribution summary

  • LLM code agents increasingly ship code with little oversight, yet defenders trained on static vulnerability datasets fall behind adapting adversaries.
  • Its 9B repository agent recovers 23 of 172 held-out VulnGym advisories, against 10 after supervised fine-tuning and 7 for the same backbone untrained under the same scaffold, and locates 17 of 43 reachable entry points against 13 for…
  • Under a judge distinct from its training oracle it cuts attack success from 36.0% to 4.6%, against WildGuard's 8.9%.

Why it matters for eval

  • LLM code agents increasingly ship code with little oversight, yet defenders trained on static vulnerability datasets fall behind adapting adversaries.
  • Under a judge distinct from its training oracle it cuts attack success from 36.0% to 4.6%, against WildGuard's 8.9%.

Researcher checklist

  • Human feedback protocol is explicit

    Detected: Red Team

  • Evaluation mode is explicit

    Detected: Automatic Metrics

  • Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Benchmark or dataset anchors are present

    No benchmark/dataset anchor extracted from abstract.

  • Metric reporting is present

    Detected: accuracy, f1