Typed states for the displayed outbound observations.
Source: paper_references, paper_reference_links, observed 2026-08-02T07:01:23.147046Z
Paper Citation Record · LEDGER
As of 18 August 2026, this Paper Citation Record lists 28 of 28 outbound references and 0 inbound Pith citation observations for arXiv:2607.13083.
A citation records a reference. It does not transfer a finding from one paper to another.
Typed states for the displayed outbound observations.
Source: paper_references, paper_reference_links, observed 2026-08-02T07:01:23.147046Z
One-hop event checks from named stored sources.
Source: scholarly_work_events, retraction_status_cache, observed 2026-08-18T06:34:40.430872+00:00
Pith citing papers itemized under the disclosed page cap.
Source: paper_references, paper_reference_links
A source-named dated measurement, never combined with another source.
Source: cited_works
28 of 28 outbound references displayed
External citation measurements
No source-named external measurement is stored.
Observation af5421f9-5c79-4bd3-909e-28d7a610d778 · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened OR-Bench: An Over-Refusal Benchmark for Large Language Models
Reference 1
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation f7ef4e9a-452b-4073-a590-901d3f6eefe4 · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Unresolved cited work
Reference 2
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation d4d35aed-bf20-45e4-8ac7-d0cb115867fd · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened SCRIBE: Structured Mid-Level Supervision for Tool-Using Language Models
Reference 3
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 4f1c9d35-4cd3-44ad-b3ff-bab706e418c4 · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Unresolved cited work
Reference 4
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 662be392-25cb-4fda-a729-5dbf7a3ac4d7 · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Meta-Harness: End-to-End Optimization of Model Harnesses
Reference 5
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation b2382074-deff-4c7a-a6d4-0d34431547a0 · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Analysing Chain of Thought Dynamics: Active Guidance or Unfaithful Post-hoc Rationalisation?
Reference 6
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 0c42fcc9-aff9-47f8-83a0-0293f9a6fefa · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Unresolved cited work
Reference 7
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 3b8334a4-eacb-4c28-8943-1f898beebaf5 · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened SafetyRepro: Configuration-Conditional Rank Instability on Alignment Benchmarks
Reference 8
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation f4a52139-6642-4601-b802-6722bdf0fd23 · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses
Reference 9
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation e5520ea9-d240-4176-85c2-40c56431da6d · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Reflect-Guard: Enhancing LLM Safeguards against Adversarial Prompts via Logical Self-Reflection
Reference 10
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 83cb1264-bb3c-481a-95f8-6be96b296c21 · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened You Cannot Fix What You Cannot Find! An Investigation of Fault Localization Bias in Benchmarking Automated Program Repair Systems
Reference 11
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 44c67161-c09e-403a-8f6a-256350610aac · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Unresolved cited work
Reference 12
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation d3862ee3-275f-4aa3-8a1b-9ae35985df42 · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Categorizing Variants of Goodhart's Law
Reference 13
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 949024c6-0e48-453d-b1ed-46b951b70c33 · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Code as Agent Harness
Reference 14
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 75082944-7ea8-4635-a2f9-65ab8894e5ce · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Evolving Agents in the Dark: Retrospective Harness Optimization via Self-Preference
Reference 15
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation b8868949-339a-43c0-835c-d4bee6e7e779 · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Unresolved cited work
Reference 16
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 27ecff90-29e8-47fd-80d3-a35eab35a43d · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened LLM Agents Already Know When to Call Tools -- Even Without Reasoning
Reference 17
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 6c8c60fb-bb4c-4cbc-bdd1-5eb025eeb94b · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Beyond Accuracy: Measuring Bias Acknowledgment in Chain-of-Thought Reasoning for Responsible AI Evaluation
Reference 18
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation af67fb16-db73-489c-a556-c8d266d60bea · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened VeRO: A Harness for Agents to Optimize Agents
Reference 19
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 4f1244d6-2d8a-4511-8359-fa9953162476 · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Unresolved cited work
Reference 20
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation f2b6b363-19e1-4878-b8b1-d4691d625625 · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Whitson and Adam D
Reference 21
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation f7eb97d0-ab92-4840-aa81-2c947bc2927b · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Adapting the Interface, Not the Model: Runtime Harness Adaptation for Deterministic LLM Agents
Reference 22
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 2325e54d-7d58-4c2b-9e4d-774778e6dacf · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened PAFT: Preservation Aware Fine-Tuning for Minimal-Edit Program Repair
Reference 23
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 5056d85c-be0b-4dd2-9a6e-b88e4fdee633 · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Unresolved cited work
Reference 24
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 0e1aaed2-53ac-4a13-8bc1-9161665f35a8 · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Reward Auditor: Inference on Reward Modeling Suitability in Real-World Perturbed Scenarios
Reference 25
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation f950cb56-c969-4987-acb1-fb549a5108ee · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened Defining and Characterizing Reward Hacking
Reference 2022
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation 856a6a25-a61c-47ea-be25-8fd08d07a997 · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened InThe Thirteenth International Conference on Learning Represen- tations
Reference 2025
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
Observation e9f73dff-2960-4a24-8cf7-a864352ab80a · outbound
Phantom Guardrails: When Self-Improving Agent Harnesses Fix Failures That Never Happened QiMeng-PRepair: Precise Code Repair via Edit-Aware Reward Optimization
Reference 2026
Source-reported events for the cited work
Unavailable: canonical work link unavailable.
No inbound Pith citation observations are available.