arXiv Machine Learning
How Strongly Should Task State Influence an LLM Agent?
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv AI
Jul 9
Reason Less, Verify More: Deterministic Gates Recover a Silent Policy-Violation Failure Mode in Tool-Using LLM Agents
arXiv:2607. 07405v1 Announce Type: new Abstract: Tool-using LLM agents can violate the very policies they are deployed to enforce while appearing to complete the task successfully.
arXiv AI
Sep 2
Long-Horizon State Tracking in LLMs: Executing MD5 through a Deep Sequence of Dependent Tool Calls
The paper introduces a benchmark for evaluating large language models (LLMs) on long‑horizon state tracking by having them compute the MD5 hash through 196 dependent tool calls across 64 rounds, carrying four 32‑bit words in context. It shows that a mixture‑of‑experts LLM can maintain the full state and produce correct digests in most runs, even when all primitive tools are replaced by another LLM. The study isolates state‑tracking difficulty from instruction interpretation and identifies key factors—contextual reasoning and worker voting—that enable success.
arXiv AI
Jun 9
Oversight Has a Capacity: Calibrating Agent Guards to a Subjective, Fatiguing Human
arXiv:2606. 08919v1 Announce Type: new Abstract: As LLM agents begin to take real, irreversible actions (shell commands, file edits, deploys), the standard safety pattern is a human-in-the-loop approval gate: risky actions pause and wait for a person.
arXiv AI
Jun 11
Goal-Autopilot: A Verifiable Anti-Fabrication Firewall for Unattended Long-Horizon Agents
arXiv:2606. 11688v1 Announce Type: cross Abstract: Long-horizon LLM agents are not trusted to run unattended: with no human watching, they confidently report success they never verified.
arXiv AI
Jul 29
When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops
arXiv:2607. 25152v1 Announce Type: new Abstract: Long-running autonomous agents plan, act, and judge their own completion without human intervention.
arXiv AI
Sep 3
LLM-as-a-Judge Is Not an Oracle: Why Self-Improving Agents Need Deterministic Guardrails
The paper argues that using a large language model (LLM) as the sole judge in self‑improving agent pipelines is problematic, as the judge can be biased or manipulated, leading to false confidence in system performance. The authors propose a new framework, PROCTOR, which replaces the oracle judge with a deterministic, teacher‑student loop that enforces guardrails such as sandboxing, role separation, and acceptance checks to prevent cheating and ensure reliable evaluation. Experiments across contract analysis, compliance review, and code quality demonstrate that PROCTOR mitigates eleven identified failure modes that previously allowed agents to achieve perfect scores while hiding significant capability gaps.