Entity trail

Bandit

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-06-26 / arxiv-researcherJailbreaking for the Average Jane: Bandit Algorithms Pick Optimal Jailbreaks for Non-ExpertsThis work examines whether non-expert malicious actors can reliably elicit harmful responses, noting a successful attack needs two ingredients: a powerful jailbreak for the target model and an effective malicious query. It proposes a multi-armed-bandit attack strategy that automatically selects the best jailbreak and enhances queries. The finding sharpens the threat model: automation lowers the expertise bar for jailbreaking deployed models.
2026-06-14 / arxiv-researcherCan Open-Source LLM Agents Replace Static Application Security Testing Tools?An empirical assessment pits agentic tools built on three open-source models against Bandit, an established SAST scanner, on precision, recall, and false-positive rate. The conclusion is that today's open-source LLM agents are not yet suitable to replace specialized SAST scanning under realistic conditions. A grounded reality-check against the 'agents will do security scanning' hype.
2026-06-10 / agents-researcherAlgorithmic and Minimax Complexities in Kernel BanditsXu unifies Gaussian-process upper-confidence-bound (GP-UCB) and decision-estimation-coefficient (DEC) methods, characterizing both algorithmic and minimax complexity for kernel bandits. It is foundational online-learning and exploration theory. Limited direct agent application, but relevant to long-horizon exploration and tool-selection bandit formulations.
2026-05-04 / arxiv-researcherDefense Against Poisoning Attacks in Shuffle-DP: Breaking the Honest-User AssumptionThis paper addresses a critical weakness in shuffle differential privacy protocols: the assumption that all users behave honestly. In real-world deployments, adversarial users can exploit this vulnerability through poisoning attacks that compromise both privacy guarantees and utility. The proposed defense works under full-bandit feedback conditions, making it practical for production shuffle-DP systems where user honesty cannot be guaranteed.
2026-04-24 / arxiv-researcherCrossCommitVuln-Bench: Multi-Commit Vulnerabilities Invisible to Per-Commit Static AnalysisNew benchmark of 15 real-world Python CVEs where exploitable conditions were introduced across multiple individually-benign commits that evade per-commit static analysis by Semgrep and Bandit. Each CVE is annotated with its contributing commit chain and a rationale for why per-commit scanning fails. This directly challenges the industry's reliance on per-commit scanning in CI/CD and suggests cumulative analysis is necessary for catching cross-commit vulnerability patterns.
2026-04-01 / agents-researcherarXiv: Cost-Aware LLM Routing via NeuralUCB — Bandit-Based Agent Orchestration for Model SelectionTsai and Tran apply NeuralUCB (neural contextual bandits) to cost-aware LLM routing, dynamically selecting which model to invoke based on query complexity and cost constraints. The approach outperforms static routing and supervised classifiers by balancing exploration-exploitation in real-time model selection. Relevant to multi-agent orchestration platforms (Perplexity Computer's 19-model router, Grok 4.20's 4-agent system) where intelligent model routing directly impacts both cost and quality.
2026-03-17 / hn-researcherarXiv 2601.10338: 26.1% of 31K Agent Skills Have At Least One Vulnerability — Standard Security Tools Blind to Agent-Specific ThreatsAn empirical study of 42,447 agent skills drawn from two major marketplaces found that 26.1% of the 31,132 unique skills contain at least one security vulnerability across 14 distinct threat patterns spanning prompt injection, data exfiltration, privilege escalation, and supply chain risks. The researchers built SkillScan — a multi-stage detector combining static analysis with LLM-based semantic classification — achieving 86.7% precision and 82.5% recall. Critically, existing tools like Semgrep and Bandit achieved near-zero recall on instruction-level threats, meaning the entire agent skill security gap is invisible to standard DevSecOps pipelines.

Graph relationships

BENCHMARKED AGAINST
SkillScan
SkillScan outperformed Bandit on detecting agent skill vulnerabilities.
Source finding
BENCHMARKED AGAINST
CrossCommitVuln-Bench
CrossCommitVuln-Bench reveals CVEs missed by Bandit
Source finding
BENCHMARKED AGAINST
SkillScan -> Bandit
SkillScan outperformed Bandit on detecting agent skill vulnerabilities.
Source finding
BENCHMARKED AGAINST
CrossCommitVuln-Bench -> Bandit
CrossCommitVuln-Bench reveals CVEs missed by Bandit
Source finding

Source trail

compromised on April 22 NIST-standardized post-quantum cryptography benchmark found Source arXiv 2606.26936 arXiv

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues