Paper Citation Record · LEDGER

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops

As of 17 August 2026, this Paper Citation Record lists 40 of 40 outbound references and 0 inbound Pith citation observations for arXiv:2607.25152.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2607.25152 v1

Coverage vector

measured 40 of 40 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-07-31T00:06:02.229517Z

measured 40 of 40 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-17T06:30:58.91139+00:00

measured 0 of 0 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: cited_works

Reference resolution

40 of 40 outbound references displayed

verified exact0
verified fuzzy0
unresolved40
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation ba35ee0c-a4db-435c-9a67-67491996ea4a · outbound

This paper cites From Confident Closing to Silent Failure: Characterizing False Success in LLM Agents.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops From Confident Closing to Silent Failure: Characterizing False Success in LLM Agents

Reference 1

Resolution: unresolved; no resolver link, observed 2026-07-31T00:05:59.121624Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1ba18b0d-4372-4b9e-9146-a5d73a93a127 · outbound

This paper cites Constitutional AI: Harmlessness from AI Feedback.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Constitutional AI: Harmlessness from AI Feedback

Reference 6

Resolution: unresolved; no resolver link, observed 2026-07-31T00:05:59.584002Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bb4a6551-b82f-4d42-a567-e2e50ff28a51 · outbound

This paper cites Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision

Reference 8

Resolution: unresolved; no resolver link, observed 2026-07-31T00:05:59.702622Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c3296c37-e76f-455f-9bad-bc9a8554d11a · outbound

This paper cites ScopeJudge: Cost-Aware Pre-Execution Gating for Offensive Security Agents.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops ScopeJudge: Cost-Aware Pre-Execution Gating for Offensive Security Agents

Reference 10

Resolution: unresolved; no resolver link, observed 2026-07-31T00:05:59.839451Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c371ef5f-2110-42c9-b13b-004fbc0ba343 · outbound

This paper cites Deep Re- inforcement Learning from Human Prefer - ences,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Deep Re- inforcement Learning from Human Prefer - ences,

Reference 11

Resolution: unresolved; no resolver link, observed 2026-07-31T00:05:59.907437Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b9e21d03-5745-4716-9552-e616413f705c · outbound

This paper cites EvilGenie: A Reward Hacking Benchmark.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops EvilGenie: A Reward Hacking Benchmark

Reference 13

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:00.011035Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation df82bd36-55ea-4ca2-a8aa-20f4bd708f46 · outbound

This paper cites Scal - ing Laws for Reward Model Overoptimiza- tion,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Scal - ing Laws for Reward Model Overoptimiza- tion,

Reference 14

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:00.086887Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9d69b39d-f819-499e-8920-194b40d65375 · outbound

This paper cites Large Language Models Cannot Self-Correct Reasoning Yet,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Large Language Models Cannot Self-Correct Reasoning Yet,

Reference 15

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:00.137455Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 912e8103-f563-4ad6-b9e9-e9a808abea03 · outbound

This paper cites AI safety via debate.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops AI safety via debate

Reference 16

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:00.192018Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7f926ccb-1d5e-4c05-9e73-e560f5900863 · outbound

This paper cites Controlled Exper- iments on the Web: Survey and Practical Guide,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Controlled Exper- iments on the Web: Survey and Practical Guide,

Reference 18

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:00.347221Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8aba963f-f187-4d10-82b5-07710c08c0d8 · outbound

This paper cites Scalable agent alignment via reward modeling: a research direction.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Scalable agent alignment via reward modeling: a research direction

Reference 21

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:00.562062Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7f0ccc12-c41a-48c8-98fb-179d494630ee · outbound

This paper cites Early Diagnosis of Wasted Computation in Multi-Agent LLM Systems via Failure-Aware Observability.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Early Diagnosis of Wasted Computation in Multi-Agent LLM Systems via Failure-Aware Observability

Reference 22

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:00.611392Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b7441adf-2310-4c39-991d-6033c8b791bc · outbound

This paper cites Let’s Verify Step by Step,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Let’s Verify Step by Step,

Reference 23

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:00.693551Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9a0517dc-6130-461b-893c-de4ab8ddb17b · outbound

This paper cites Retrospec- tive Progress-Aware Self-Refinement for LLM Agent Training,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Retrospec- tive Progress-Aware Self-Refinement for LLM Agent Training,

Reference 24

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:00.752157Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation effd6ed5-f217-46d1-9037-895159e7c9f4 · outbound

This paper cites Self-Refine: Iterative Refinement with Self-Feedback,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Self-Refine: Iterative Refinement with Self-Feedback,

Reference 25

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:00.814656Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 21f9bc9a-3029-432d-8d7d-9d4c6c2dffde · outbound

This paper cites Categorizing Variants of Goodhart's Law.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Categorizing Variants of Goodhart's Law

Reference 26

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:00.865901Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c60514e2-ca06-4acd-b210-8d964cfcb204 · outbound

This paper cites Training Language Mod- els to Follow Instructions with Human Feedback,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Training Language Mod- els to Follow Instructions with Human Feedback,

Reference 27

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:00.931998Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4b084f1d-6454-4fbe-bb3b-74e0c7ff3cf0 · outbound

This paper cites The Effects of Reward Misspecification: Map - ping and Mitigating Misaligned Models,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops The Effects of Reward Misspecification: Map - ping and Mitigating Misaligned Models,

Reference 28

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:01.033058Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0e0dc9f1-da99-491f-9b26-d10498874b10 · outbound

This paper cites Spontaneous Reward Hacking in Iterative Self-Refinement.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Spontaneous Reward Hacking in Iterative Self-Refinement

Reference 29

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:01.103945Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 06b0bd32-9eb6-4620-8bb7-18cbf470cd0a · outbound

This paper cites LLM Evaluators Recognize and Fa- vor Their Own Generations,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops LLM Evaluators Recognize and Fa- vor Their Own Generations,

Reference 30

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:01.201590Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 37e51bc7-4821-4415-a6bb-a23c9b3c4d78 · outbound

This paper cites Reflexion: Language Agents with Verbal Reinforcement Learning,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Reflexion: Language Agents with Verbal Reinforcement Learning,

Reference 31

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:01.307077Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e5889ad6-d8e6-4686-bbd1-cc028bf4d6d3 · outbound

This paper cites Defining and Char- acterizing Reward Hacking,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Defining and Char- acterizing Reward Hacking,

Reference 32

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:01.395358Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 300eacf1-8137-437f-891e-c2b8b8d2333a · outbound

This paper cites Solving math word problems with process- and outcome-based feedback.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Solving math word problems with process- and outcome-based feedback

Reference 33

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:01.464945Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 06b51709-4ae3-4445-b4ac-b6c889540733 · outbound

This paper cites The Verification Horizon: No Silver Bullet for Coding Agent Rewards.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops The Verification Horizon: No Silver Bullet for Coding Agent Rewards

Reference 34

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:01.599726Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4cac049e-4bbc-4de3-acb2-ad957a09c817 · outbound

This paper cites Forage V2: Knowledge Evolution and Transfer in Autonomous Agent Organizations.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Forage V2: Knowledge Evolution and Transfer in Autonomous Agent Organizations

Reference 35

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:01.663729Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 12ca15e8-2dee-4054-b2cd-3432047aef7e · outbound

This paper cites SWE-agent: Agent- Computer Interfaces Enable Automated Software Engineering,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops SWE-agent: Agent- Computer Interfaces Enable Automated Software Engineering,

Reference 36

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:01.801069Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ce56bc70-e032-48e9-b970-c352c3180329 · outbound

This paper cites ReAct: Syn- ergizing Reasoning and Acting in Language Models,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops ReAct: Syn- ergizing Reasoning and Acting in Language Models,

Reference 37

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:01.918907Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4bd76697-74f8-4787-b2da-f22913978e9a · outbound

This paper cites $\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops $\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains

Reference 38

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:02.005364Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 46f31ed6-b2f3-450c-81d9-0cda687c4198 · outbound

This paper cites SpecBench: Measuring Reward Hacking in Long-Horizon Coding Agents.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops SpecBench: Measuring Reward Hacking in Long-Horizon Coding Agents

Reference 39

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:02.127727Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3e525af4-cf19-4908-a32b-a58f072f4bd4 · outbound

This paper cites Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena,

Reference 40

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:02.229517Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bb047815-d530-49fb-9be7-bb482fd10599 · outbound

This paper cites Kohavi, D.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Kohavi, D

Reference 2009

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:00.404153Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5a7bf308-a0e1-4d9b-9ac3-1e8ecb116fa5 · outbound

This paper cites Building Effective AI Agents,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Building Effective AI Agents,

Reference 2016

Resolution: unresolved; no resolver link, observed 2026-07-31T00:05:59.408979Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ceaeaecc-b479-4e83-98fa-9f6965791359 · outbound

This paper cites GroundEval: A Deterministic Replacement for LLM-as-Judge in Stateful Agent Evaluation.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops GroundEval: A Deterministic Replacement for LLM-as-Judge in Stateful Agent Evaluation

Reference 2017

Resolution: unresolved; no resolver link, observed 2026-07-31T00:05:59.959934Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dbb15f01-af43-4cde-b83f-623c58ad06e1 · outbound

This paper cites SWE-bench: Can Language Models Resolve Real-World GitHub Issues?,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops SWE-bench: Can Language Models Resolve Real-World GitHub Issues?,

Reference 2018

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:00.294890Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9de3934e-92c4-4e34-bcee-718312eaf032 · outbound

This paper cites Specification Gaming: The Flip Side of AI Ingenuity,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Specification Gaming: The Flip Side of AI Ingenuity,

Reference 2020

Resolution: unresolved; no resolver link, observed 2026-07-31T00:06:00.478706Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a7683982-5cd8-4534-b471-02c2786afd19 · outbound

This paper cites Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-07-31T00:05:59.643757Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 290563ce-728e-4d18-917d-dcad6186e0c5 · outbound

This paper cites Reward Hack- ing in Language Model Agents: Revisiting AI Safety Gridworlds,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Reward Hack- ing in Language Model Agents: Revisiting AI Safety Gridworlds,

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-07-31T00:05:59.783791Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 27e3f52e-3632-40c0-a007-5f08b151d0ac · outbound

This paper cites Effective Harnesses for Long- Running Agents,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Effective Harnesses for Long- Running Agents,

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-07-31T00:05:59.450472Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d1030f6f-d2ce-434f-9943-3b5aac86714d · outbound

This paper cites Reward - HackingAgents: Benchmarking Evalua - tion Integrity for LLM ML-Engineering Agents,.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Reward - HackingAgents: Benchmarking Evalua - tion Integrity for LLM ML-Engineering Agents,

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-07-31T00:05:59.528265Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 13b1f4b6-1c8e-4072-b87b-7f93880075d9 · outbound

This paper cites Concrete Problems in AI Safety.

When Do Agent Loops Mistake Stagnation for Progress? Self-Evaluation Bias and Externally Grounded Verification in Long-Running Autonomous LLM Agent Loops Concrete Problems in AI Safety

Reference 2026

Resolution: unresolved; no resolver link, observed 2026-07-31T00:05:59.249687Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.