2024-06-19T08:55:56Z
AgentDojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents
Selected methodological reference
Reading desk
Read the work.
Question the method.
Selected primary research relevant to agents, evaluation, context and practical deployment. Paper metadata and links are distinct from our own findings. arXiv hosts preprints; listing a paper is not endorsement or peer-review verification.
2025-03-17T19:04:38Z
Why Do Multi-Agent LLM Systems Fail?
Selected methodological reference
2026-02-23T13:47:43Z
MAS-FIRE: Fault Injection and Reliability Evaluation for LLM-Based Multi-Agent Systems
Selected methodological reference
2024-06-17T19:33:08Z
$τ$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains
Selected methodological reference
2026-09-04T01:23:47Z
$τ^τ$-Bench: An Environment for End-To-End, Realistic Agent Construction
Selected methodological reference
2026-10-15T17:56:07Z
3DCodeVerse: Exploring 3D Worlds as Code
Automated discovery; editorial review pending
2026-10-15T17:51:57Z
Learning Self-Orchestration in Decentralized Agent Swarms
Automated discovery; editorial review pending
2026-10-15T17:51:32Z
A team of Gemini 4 Argon agents beats Minecraft
Automated discovery; editorial review pending
2026-10-15T17:51:17Z
Learning to Steer LLM Agents and Multi-Agent Systems
Automated discovery; editorial review pending
2026-10-15T17:43:47Z
CAT-alyzing Honesty: Continuous Adversarial Training for Transferable LLM Honesty
Automated discovery; editorial review pending
2026-10-15T16:45:59Z
HTN-Orchestrate: A Hierarchical Task Network Framework for Multi-Agent Orchestration
Automated discovery; editorial review pending
2026-10-15T16:44:34Z
AutoSFT: An Agentic System for Automated Data Research in Supervised Fine-Tuning
Automated discovery; editorial review pending
2026-10-15T16:12:07Z
Believed but Not Said: Measuring Hidden Belief Collapse in Multi-Agent LLM Discussions
Automated discovery; editorial review pending
2026-10-15T16:01:38Z
PROBE: Learning to Audit Policy Compliance in Tool-Using LLM Agents
Automated discovery; editorial review pending
2026-10-15T15:12:56Z
WEAVE: Weaving Real-World Code Evolution into Dependency-Structured Tasks for Benchmarking and Training Code Agents
Automated discovery; editorial review pending
2026-10-15T14:00:33Z
CodeHealth Predicts the Token Cost of Agentic Code Review: A Pre-Registered Study of 465 Agent-Authored Pull Requests
Automated discovery; editorial review pending
2026-10-15T13:59:45Z
Learning Rewards in Strategic Games
Automated discovery; editorial review pending
Daily source monitor
Newly retrieved
paper metadata.
Automated arXiv discovery. These records have not been editorially reviewed; they are a reading queue, not MEKreflect findings.
Loading the last successful source snapshot…