Search: “agent-evaluation”

Reflectiz Launches Agentic Pentesting for Websites: Up to 10x Coverage vs Conventional Pentests

Reflectiz released an agentic penetration testing platform using specialized AI agents to discover, attack, and validate web vulnerabilities with independently validated findings.

Reflectiz announced a multi-agent penetration testing capability for websites, part of its new Offensive Hub alongside Security Hub and Privacy Hub. Four agent roles map, fingerprint, attack, and validate web applications, with claims of up to 10x coverage over conventional pentesting due to pre-existing site models. Results are delivered with reproduction steps and coverage maps, and integrate via REST API, CI/CD triggers, and Slack. This syndicated CyberNewswire press release duplicates the vendor announcement.

Cyber Security News·Sep 8, 14:02 UTC · 7d agoTools 2 sources

ExecCritic: Learn to Test, Test to Improve for Coding Agents

ExecCritic separates test generation from patching for coding agents, lifting SWE-bench Verified resolution to 72.6%.

ExecCritic pairs a test-verify-revise scaffold with role-specific reinforcement learning: a Test agent writes repository-native tests and a Repair agent fixes code from execution feedback, both using Qwen-3.5-35B-A3B backbones. Post-trained Qwen agents compose to 72.6% on SWE-bench Verified, an 11.4-point gain over the 61.2% no-test baseline, without stronger-model or oracle feedback at evaluation time. The work shows test quality is the key variable: base-agent tests lowered resolution to 57.3% while GPT-5.6-sol tests raised it to 65.3%.

arXiv cs.AI / cs.LG / cs.CL·Sep 8, 17:53 UTC · 7d agoAI research1

AgentGrad: Intervention-guided Prompt Optimization for Multi Agent Systems

AgentGrad introduces intervention-guided prompt optimization for LLM multi-agent systems, achieving state-of-the-art results with 2.5x faster optimization.

AgentGrad is a prompt optimization framework for LLM-based multi-agent systems that addresses limitations in textual gradient extraction and aggregation. It uses sequential intervention to identify the agent whose prompt modification resolves a given failure, then applies agent-level supervision and semantic gradient clustering to build generalized gradients. Experiments report state-of-the-art performance across five MAS benchmarks and a 2.5x average reduction in wall-clock optimization time versus the next-fastest baseline.

Hugging Face daily papers·Sep 7, 20:00 UTC · 8d agoAI research

How well do agents use test/verification techniques?

Dan Luu's eval finds coding-agent testing instructions (TDD, formal methods, PBT, skills) mostly fail to beat defaults on Zstd implementation correctness.

The author ran 26 prompt conditions plus 4 skills on a Zstd-in-Rust implementation eval using codex with GPT-5.6, testing TDD, fuzzing, property-based testing, formal methods (Lean 4, TLA+, Verus, Kani, SMT solvers) and community skills. Nothing dramatically outperformed the default no-instruction condition, which did above average; at xhigh effort, fuzzing and PBT conditions did slightly better than formal methods. Pre-registered predictions included TDD underperforming and popular test skills (ECC, Hegel, Trail of Bits) not outperforming. Results are averages of 80 runs per condition plotted against cost.

Hacker News · security·Sep 8, 02:58 UTC · 8d agoAI researchHN 131↑ · 50 comments

What Does an LLM-Agent Leaderboard Rank Actually Compare?

A methodological study shows close LLM-agent leaderboard rank gaps on SWE-bench and similar benchmarks often do not support superiority claims.

The paper defines an estimand-aware pairwise procedure for comparing agents, checking common support and applying explicit uncertainty rules and practical margins. Across SWE-bench, AgentRewardBench, and tau2-bench, close rank differences are frequently unresolved, and proxy labels or utility rules can change which system is selected. The authors argue a leaderboard score summarizes a released evaluation but does not by itself justify pairwise superiority conclusions.

arXiv cs.AI / cs.LG / cs.CL·Sep 7, 17:31 UTC · 8d agoAI research

When an Attacker Meets a Group of Agents: Navigating Amazon Bedrock's Multi

Unit 42 red-teamed Amazon Bedrock multi-agent applications, demonstrating prompt-injection attack chains that leak agent instructions and invoke tools, mitigated by Bedrock Guardrails.

Unit 42 red-teamed Amazon Bedrock Agents' multi-agent collaboration in Supervisor and Supervisor with Routing modes. The demonstrated attack chain detects the operating mode, discovers collaborator agents, delivers attacker-controlled payloads, and can disclose agent instructions and tool schemas and invoke tools with attacker-supplied inputs. No vulnerabilities were found in Bedrock itself, and the built-in prompt attack Guardrail blocked the attacks when properly configured. The researchers collaborated with Amazon's security team and frame the findings as a broader prompt injection risk for LLM-based systems.

Palo Alto Unit 42·Aug 17, 12:31 UTC · Aug 17, 2026AI safety & security

The Intelligible World of Agents

Recorded Future argues cybersecurity AI agents perform better when reasoning over structured, curated intelligence graphs rather than fragmented alerts or open-source noise.

In a vendor essay, Recorded Future describes how its security agents produced more authoritative analyses after being re-architected to reason primarily over the Recorded Future Intelligence Graph instead of weighting open-source information equally. The author argues agentic decision quality depends mainly on a structured, current operational world model of assets, vulnerabilities, threat actors, detections and organizational context, not on model intelligence itself. The piece further claims frontier model access is commoditizing and that orchestration tooling will converge, making trusted representations of organizational knowledge the durable competitive differentiator.

Recorded Future·Sep 10, 00:00 UTC · 6d agoAI safety & security

SAEScientist-Bench: Can AI Agents Conduct Autonomous SAE Interpretability Research?

SAEScientist-Bench evaluates whether AI agents can autonomously conduct SAE interpretability research in Gemma-2-9B-IT, finding frontier agents trail expert baselines.

SAEScientist-Bench tests if AI agents can act as scientists using SAE tools for autonomous mechanistic discovery, requiring them to design contrastive probes and navigate a Gemma Scope dictionary of 131K+ features in Gemma-2-9B-IT. Across 10 agent configurations and 20 tasks, frontier agents showed genuine discovery capability but remained well behind expert reference features, lagging most in causal steering. Agents frequently misinterpreted experimental measurements even when designing effective contrasts.

Hugging Face daily papers·Sep 7, 20:00 UTC · 8d agoAI research

SAEScientist-Bench: Can AI Agents Conduct Autonomous SAE Interpretability Research?

SAEScientist-Bench tests whether AI agents can autonomously run SAE interpretability research on Gemma-2-9B-IT; frontier agents trail expert baselines.

The benchmark requires agents to design contrastive probes and navigate the Gemma Scope dictionary of over 131K features in Gemma-2-9B-IT to discover optimal interpretable features, scored against expert-curated references on Neuronpedia via activation rank, concept selectivity, and causal steering. Across 10 agent configurations and 20 tasks, frontier agents demonstrate genuine discovery capability and approach expert levels at separating target concepts from controls, but lag substantially in causal steering and frequently misinterpret experimental measurements. The authors frame this as establishing experimental model understanding as a measurable capability for closed-loop autonomous AI R&D and post-hoc monitoring for recursive self-improvement.

arXiv cs.AI / cs.LG / cs.CL·Sep 8, 17:45 UTC · 7d agoAI research1

Staying Ahead of Adversarial AI Through Agentic Source Code Review

Google Threat Intelligence details an agentic AI pipeline with human expert oversight to review source code and outpace AI-enabled attackers.

Google Threat Intelligence researchers argue that adversaries' misuse of AI raises the risk of data theft and extortion when proprietary source code is exposed. They describe a structured agentic source code review pipeline that combines AI models with skeptical validation steps and injected human domain expertise. The team reports a leap in efficacy in finding vulnerabilities before adversaries can exploit them.

Google Threat Intelligence·Aug 18, 14:00 UTC · 28d agoResearch1

An Evidence Model for Agentic Processes: Evidence Claims, Trust Assumptions, and Policy Assessment

Researchers propose an evidence claim model defining which trust and audit claims agentic AI systems can support, mapping claims to mechanisms, assumptions, and threats.

The paper proposes an evidence claim model for agentic AI processes that exchange messages, invoke tools, request approvals, and modify shared artifacts. It distinguishes claim types such as artifact integrity, provenance, approval evidence, and policy assessment, mapping each to mechanisms, assumptions, limitations, and threats. It stresses that hashes, signatures, and external anchors do not establish semantic truth, authorization, or capture completeness. The contribution is conceptual, offering vocabulary for what an agentic black box can and cannot evidence and which controls must surround it.

arXiv cs.CR·Sep 8, 09:25 UTC · 8d agoAI safety & security

When Intelligence Becomes Agency: A Theory of Governed, Proactive Agency for Symbiotic AI Systems

Researchers propose a formal framework for governed proactive AI agency, linking activation decisions to continuing authorization, accountability, and bounded delegation in symbiotic systems.

The paper defines the activation problem: persistent AI assistants must decide whether a situation warrants behavior, and whether to act, ask, monitor, defer, or deliberately refrain. It distinguishes autonomous from delegated agency and defines symbiotic agency as delegation under a standing, revocable mandate, coupled to the principal's situation with calibrated inference and bounded personalization. The framework links activation decisions to authorized perception, behavior selection, authority containment, traceable restraint, and constrained adaptation, and provides an agency classification method, evaluation framework, benchmark scenarios, and reference architecture for always-present assistants and embodied support systems.

arXiv cs.AI / cs.LG / cs.CL·Sep 7, 16:46 UTC · 8d agoAI safety & security

NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness

NeoHorse-1 introduces agentic post-training with intelligent routing that lifts agent benchmark scores at 4B and 9B scales, prototyping recursive self-improvement.

NeoHorse-1 is a family of agent-native models trained through agentic post-training: routing-harness logs (predicted capability demand, service tier, interaction) become structurally validated training data organized into a three-stage SFT curriculum plus routing-guided on-policy distillation. Capability-guided allocation converts evaluation feedback into the next training mixture, closing an evaluation-selection-update loop. Post-training raises the macro-average from 58.94 to 64.87 at 4B and from 65.60 to 69.04 at 9B across eleven agent, tool-use, coding, and instruction-following benchmarks. The authors position it as a prototype of harness-mediated recursive self-improvement.

Hugging Face daily papers·Sep 7, 20:00 UTC · 8d agoAI research

When Agents Slow Down: Understanding LLM Agents' Test-Time Strategies via Elo-per-token Analysis

Elo-per-token analysis shows LLM agents' marginal gains drop below independent sampling at scale; parallel sessions beat one long session.

The paper proposes Elo-per-token analysis, using a Bradley-Terry model to measure how agent performance scales with token budget on open-ended tasks with continuous scoring. Across four agents and four benchmarks with sessions up to 100M tokens, agents initially convert tokens to Elo faster than independent sampling but eventually slow below the linear-in-log-compute reference. The authors define a scaling inflection point and show that splitting 100M tokens across parallel sessions on FrontierCS Polyomino Packing gains +264 Elo over one long session and +355 over ten short sessions. Human contestants on shared AtCoder Heuristic Contest tasks improve superlinearly, indicating headroom over current agents.

Hugging Face daily papers·Sep 13, 20:00 UTC · 2d agoAI research3· 2 reads

τ^τ-Bench: An Environment for End-To-End, Realistic Agent Construction

New τ^τ-bench tasks coding agents with building deployable customer-service agents; best config, Claude Opus 5, passes only 23.9% of simulations.

Researchers introduce τ^τ-bench, an end-to-end benchmark where a developer agent must build a complete customer-service agent from real business records, a client with requirements, a production API, an inherited codebase, and cost/model limits, then is scored by deploying it against held-out simulated users. Across 53 tasks in four domains, the strongest configuration, Claude Opus 5 under Claude Code, passes just 23.9% of evaluation simulations versus an 82.2% expert-authored reference ceiling. Failure modes mirror those of human developers: shallow queries instead of deep record comprehension, almost no client communication, and shipping the first architecture that runs rather than experimenting.

Hugging Face daily papers·Sep 3, 20:00 UTC · 12d agoAI research

AI agent deception moves from theory to reality in UK cyber tests

Help Net Security·Aug 5, 00:00 UTC · Aug 5, 2026Data breach

Rethinking Indirect Prompt Injection as a Test-Time Search Problem

Researchers frame indirect prompt injection as test-time search, showing added attacker compute improves vulnerability discovery and exploitation against tool-using agents.

The paper models indirect prompt injection as a test-time search over a task-dependent attack surface shaped by the environment, user task, and injection goal. The authors build an agentic attacker with a dedicated search harness that performs reconnaissance, structured strategy reasoning, and adaptive evaluation using victim-agent feedback. Experiments show more attacker test-time compute improves discovery and exploitation of injection vulnerabilities, with explicit strategy management needed to avoid redundant search. The results argue that agentic security evaluations should characterize attacker search procedures and compute budgets rather than treating attack success as budget-independent.

arXiv cs.CR·Sep 3, 21:28 UTC · 12d agoAI safety & security

A Cyber Range Evaluation of Autonomous Network Incident Response Agents

Cyber range evaluation shows reinforcement learning incident response agents defend emulated networks more efficiently than heuristic policies, depending heavily on adversary behavior.

The paper evaluates agents for automated network intrusion response in a cyber range designed for human operator training, featuring variable topology, red-team emulation, and simulated users. Alerts are generated by a SIEM platform and mapped to a data modeling language used by the agents, with reinforcement learning policies optimized to minimize combined defense and availability costs using a cyber attack simulator. Reinforcement learning agents defended the system more efficiently than heuristic policies, with performance highly dependent on the adversary policy and simulated user behavior.

arXiv cs.CR·Sep 15, 02:36 UTC · 1d agoResearch

Agent as Policy for Robotic Manipulation

Agent as Policy lets a general-purpose agent drive a physical robot via runtime reasoning and program generation, reaching 100% success on manipulation tasks.

The paper introduces Agent as Policy (AGP), which puts task planning and execution for a physical robot under a general-purpose agent's control with no task-specific or environment-specific training. The agent interprets visual evidence, writes executable programs, issues motion commands, and revises actions based on physical outcomes. AGP was evaluated on real-world manipulation tasks including assembly from human videos, block construction from goal images, die reorientation, targeted throwing, and bimanual towel folding. It achieved success rates of 100%, 100%, and 80% on three block construction configurations.

Hugging Face daily papers·Sep 10, 20:00 UTC · 5d agoAI research

BenchShield: Formal Model-Backed Instrumentation for Reward Integrity in LLM-Agent Evaluation Infrastructure

BenchShield uses lifecycle-model-backed instrumentation to detect reward hacking in LLM-agent benchmarks, lifting full-chain recall to 77-100% at up to 65% lower cost.

The framework grounds reward-hacking detection in a finite lifecycle model of an evaluation's reward-relevant events, combining a static phase-aware taint analysis with runtime infrastructure-side evidence attribution. Evaluation used a human-labeled corpus of 456 adjudicated trajectories drawn from more than 31,000 public agent runs across three benchmarks. BenchShield improves full-chain recall from 23-94% to 77-100% and same-vector coverage from 16-56% to 43-78%, cuts per-task cost by up to 65%, and achieves 96% accuracy detecting reward hacking at runtime.

arXiv cs.CR·Sep 10, 03:10 UTC · 6d agoAI safety & security1

Why transparent AI agents matter more than you think

CyberScoop·Aug 10, 14:23 UTC · Aug 10, 2026Exploit / PoC in the wild1

The Agentic SOC – From AI Theater to Real Defense

Recorded Future and Accenture experts outline how security teams can move beyond 'AI theater' toward agentic SOC operations guided by measurable KPIs.

Recorded Future published a blog featuring perspectives from its own and Accenture experts on building an agentic security operations center. The piece argues organizations should prioritize measurable KPIs and proactively mitigate risks from autonomous agents. It also discusses evolving the analyst role from managing alerts to managing agents.

Recorded Future·Sep 1, 00:00 UTC · 15d agoIndustry

Reflectiz Launches Agentic Pentesting for Websites: Up to 10x Coverage vs Conventional Pentests

Reflectiz launched a multi-agent AI penetration testing platform for websites, claiming up to 10x more coverage than conventional pentests.

Reflectiz, a continuous web exposure management vendor, launched a multi-agent penetration testing platform for websites under a new Offensive Hub. Four specialized agents crawl the site like a real user, fingerprint the stack, run attacks, and independently validate findings to remove false positives. The company claims up to 10x more coverage than conventional pentesting by starting from its pre-existing live model of each site. Findings integrate via REST API, CI/CD triggers and Slack, and the Atlas remediation agent guides fixes.

CSO Online·Sep 8, 14:41 UTC · 7d agoTools

AgenticGen: Reward-Guided Agentic Video Generation for Advertising

AgenticGen applies DPO and GRPO reward-guided reasoning to ad video generation, improving TikTok CTR 2.72%, CVR 2.63%, and Advv 9.61%.

AgenticGen decomposes advertising video generation into two trainable reasoning stages, strategy selection and draft generation, supervised by online business feedback. It learns a performance-based reward from accumulated online feedback plus a rubric-based reward aligned with human quality standards, then optimizes policies with DPO followed by GRPO using process and outcome rewards. Online A/B experiments in the TikTok advertising system show CTR up 2.72%, CVR up 2.63%, and Advv up 9.61% over an SFT baseline.

Hugging Face daily papers·Aug 30, 20:00 UTC · 16d agoAI research

Hackers Weaponize Agentic AI to Automate Reconnaissance, Exploitation and Post-Exploitation

Google GTIG reports threat actors using agentic AI to automate reconnaissance, exploit selection, and credential harvesting, compromising thousands of secrets.

Google Threat Intelligence Group's Q3 2026 AI Threat Tracker documents threat actors operationalizing agentic AI: in one Mandiant investigation, a financially motivated actor built and executed a credential-harvesting operation in under six hours, with an exposed 'Recon' framework managing more than 23,800 harvested secrets including cloud and AI-service API keys. A Chinese-speaking actor tracked as knaithe used a DeepSeek-powered Hermes Agent for automated reconnaissance and vulnerability enumeration, pivoting from Langflow to n8n and enabling manual exploitation of exposed Citrix NetScaler, Marimo, Apache Tomcat, and VPN infrastructure. Operators harvested Citrix session cookies from process memory to bypass MFA, obtained AWS credentials from compromised Marimo instances, and deployed the Go-based NKAbuse backdoor, with reported RCE and data exfiltration. Google notes fully autonomous end-to-end AI attack pipelines have not yet been observed in the wild.

GBHackers·Sep 9, 09:12 UTC · 7d agoThreat actor in the wild2

EmbodiedSkills: A Unified Framework for Orchestrating, Training, and Deploying VLA Agents

Researchers introduce EmbodiedSkills, a framework treating VLA skill decisions as verified execution proposals, reaching 86.2% success on RoboTwin 2.0.

The EmbodiedSkills framework treats each vision-language-action skill decision as an execution proposal, checking prerequisites before execution and verifying outcomes afterward via a shared executable-skill interface. It connects high-level skill selection, bounded low-level VLA execution and post-action verification in a single agent loop, and logs structured trajectories for supervision and optional online adaptation. Instantiated with Qwen3-VL and OpenPI/pi0.5, task-adapted policies achieve 86.20% average success across 50 RoboTwin 2.0 tasks and 97.40% across the four LIBERO suites, with 12.5% on memory-dependent RMBench tasks.

Hugging Face daily papers·Aug 31, 20:00 UTC · 15d agoAI research1

Misleading the Planner through Deceptive Resumes: Registration-Time Injection in Centralized Multi-Agent Systems

Researchers demonstrate registration-time prompt injection in centralized LLM multi-agent systems, dropping GAIA task success from 84.31% to 37.25%, and propose DescGuard defense.

The paper identifies a registration-time injection channel in centralized LLM multi-agent systems where third-party worker agent descriptions are trusted by the planner before any user instruction arrives. Analyzing 32,000 descriptions from three public agent marketplaces, at least 23.35% contain content outside the four defined description fields. Eight description-manipulation attack strategies targeting task decomposition, capability grounding, and subtask specification cut GAIA task success from 84.31% to 37.25% and increased token consumption or execution time by over 111%, persisting across two MAS implementations, six planner LLMs, and four evaluators. The proposed DescGuard defense filters descriptions to worker-scoped interface information and restores metrics toward baseline without modifying workers, planner, or orchestration logic.

arXiv cs.CR·Sep 14, 13:06 UTC · 2d agoAI safety & security

Verifiable Social Reasoning for LLM Assistants

Fuse, a multi-agent simulation with hidden motives, evaluates LLM social reasoning, revealing compounding difficulty from user mediation and bias sensitivity.

Fuse is a multi-agent simulation framework in which a target agent with a hidden motive interacts with other agents including one representing the user, who consults the evaluated assistant to infer the motive, providing verifiable ground truth by construction. Simulation faithfulness is validated through a human study with 24k annotations. Applied to 12 LLMs, it shows user mediation compounds social reasoning difficulty, models are systematically sensitive to biased user framing, models may need more details than humans, and longer conversations do not always improve performance. The framework and a 21k-example dataset are open-sourced.

arXiv cs.AI / cs.LG / cs.CL·Sep 15, 17:37 UTC · 19h agoAI research

Empirical Evaluation of Task-Based Permission Scoping Architecture for AI Agents

Fine-tuned RoBERTa-large task permission classifier matches Claude Haiku 4.5 on access scoping for AI agents, cutting severity-weighted attack surface by 84.4%.

The paper evaluates a three-source task-based permission architecture for AI agents combining role-based permission ceilings, a task permission classifier, and policy-based prohibitions. A fine-tuned RoBERTa-large security gate matched few-shot Claude Haiku 4.5 on a 600-prompt dataset, with macro-F1 0.881 versus 0.886, precision 0.897 versus 0.842, and lower severity-weighted residual risk (0.63 versus 1.12). An attack-surface elimination metric shows the role ceiling alone closes 27.9% of the severity-weighted surface while adding the task classifier closes 84.4%. The work establishes task-granular access control as a measured, deployable mechanism for reducing attack surface in agentic deployments.

arXiv cs.CR·Sep 14, 11:48 UTC · 2d agoAI safety & security