arXiv AI

CARGO: Context-Aware Retrieval-Gated Evaluation of Agentic AI in Production

CARGO is a framework for evaluating agentic AI systems in production that addresses the problem of reference-instance divergence (RID), where reference-based judges penalize correct answers that involve different entity identifiers. It treats retrieved references as procedural exemplars, grounds judgments in the live instance’s context, assigns a three-way status to claims, and gates evaluation by retrieval confidence. Using the CARGO-Bench diagnostic suite, CARGO eliminates false penalties and improves discrimination while revealing a limitation in detecting procedural corruptions.

arXiv Computation and Language
Oct 22

Conflicted Records: Typed Adversarial Contracts for Auditable LLM Output Quality Enforcement

The paper introduces BiasGuard‑D, a system that uses two adversarial language‑model agents from different families to exchange typed claim documents, with a third model quorum voting to adjudicate. A combined referential‑plus‑semantic invariant validates the verdict, and the iterative deliberation layer shifts remediation from stereotypes to hallucinations, improving right‑answer recovery by 49 percentage points on HaluEval QA. All claims, votes, and instructions are stored as typed records, providing an auditable provenance trail for operators under emerging AI regulations.

By Kardelen Subasi, Jaros{\l}aw A. Chudziak
arXiv AI
Sep 2

trajectory-judge: What Outcome-Only LLM Judges Miss on Agent Trajectories

The paper "trajectory-judge: What Outcome-Only LLM Judges Miss on Agent Trajectories" examines the limitations of outcome-only evaluation for large language model agents. Using a deterministic tool‑using support‑desk environment with a scripted oracle policy and a fault injector, the authors compare five different judging approaches—programmatic rules, outcome‑only, step‑rubric at two model sizes, and a self‑consistency ensemble—on metrics such as detection, step localisation, fault typing, calibration, and cost across 400 trajectories. The study finds that outcome‑only judges miss many silent faults and generate false positives, while step‑rubric judges achieve higher recall with no false alarms but at greater cost, and that none of the judges read the final reply, allowing fabricated promises to evade detection. "whyItMatters":"The findings highlight that current production‑default outcome‑only evaluations can overlook critical failures in agent behavior, underscoring the need for more nuanced, step‑level judging methods to ensure reliable LLM agent performance."

By Hadi Mohammadi
arXiv AI
2d ago

DISCERN: Preserve at Write, Resolve at Read for Persistent LLM-Agent Memory

The paper introduces DISCERN, a memory framework for large language‑model agents that preserves information at write time and resolves ambiguities at read time. By constructing a bi‑temporal provenance graph that never merges confusable entities or drops superseded values, DISCERN enables a zero‑LLM tool to navigate the graph and make calibrated decisions about answering or abstaining. On the ENT‑310 benchmark, DISCERN outperforms existing memory systems in both abstention calibration and macro answer correctness, largely due to its superior entity‑resolution capability.

By Chirag Ahuja, Vladislav Tanov, Ramcharan Kakarla, Jonathan Taws
arXiv AI
Aug 24

No Judgment Without a Reason: Counterfactual Receipts for Versioned AI Evaluators

The paper introduces a framework for evaluating AI systems that not only checks final labels but also tracks the reasoning behind them through three core sources—grounds, norms, and authority—forming an eight-cell counterfactual judgment cube. It defines minimal source replacement sets, called judgment receipts, to explain changes in verdicts and provides certification cost bounds for black-box evaluators. The authors present ReasonBench, a benchmark with 19,520 cases, and demonstrate that while high standard accuracy can mask robustness issues, receipt accuracy reveals significant gaps in reasoning consistency across different models.

By Ye Chen, Weining Zhang
arXiv AI
Jul 3

ContextNest: Verifiable Context Governance for Autonomous AI Agent

arXiv:2607. 02116v1 Announce Type: new Abstract: Autonomous AI agents increasingly depend on external knowledge stores, yet most retrieval pipelines provide relevance without durable guarantees of provenance, version identity, integrity, traceability, or point-in-time reconstruction.

By Misha Sulpovar (PromptOwl, LLC), Benn R. Konsynski (Goizueta Business School, Emory University), Qaish Kanchwala (IBM Research), Gabe Goodhart (IBM Research)
arXiv AI
Nov 6

Are LLM Safety Judges Policy-Invariant? A Three-Principle Stress-Test

The paper investigates whether large language model (LLM) safety judges are truly policy‑invariant, meaning their verdicts should depend only on the agent’s behavior and not on how the evaluation policy is phrased. It introduces three testable principles—rubric‑semantics invariance, rubric‑threshold invariance, and ambiguity‑aware calibration—and applies them in a stress‑test protocol across four agent‑class judges using trajectories from ASSEBench and R‑Judge. The study finds that current judges often change verdicts in response to both meaningful normative shifts and meaningless structural rewrites, with up to 9.1% of verdicts flipping due to content‑preserving rewrites and 18‑43% of flips occurring on unambiguous cases, indicating that safety scores conflate agent actions with evaluator wording. whyItMatters":"The findings reveal a hidden failure mode in safety evaluation that could mislead researchers about agent performance, underscoring the need for policy‑invariant judges and transparent reporting protocols."

By Shihao Weng, Yang Feng, Xiaofei Xie
arXiv Computer Vision
Sep 4

VeriPhy: Agentic Physical Reasoning for World Model Evaluation and Refinement

VeriPhy is an auditable physical‑verification system that transforms a text prompt into typed physical obligations and a statically validated execution plan before any video frames are generated. During execution, it gates calls to frozen low‑level experts (segmentation, tracking, counting, depth, OCR, audio‑event detection, etc.) and records provenance‑carrying evidence for each action. The system maps these records to a three‑valued state—supported, contradicted, or unknown—providing traceable verdicts that can be used to refine generation models.

By Wenzhuo Xu, Yuchen Zhu, Chongjian Ge, Xuan Shen, Jing Shi, Jason Kuen, Yongxin Chen, Molei Tao, Christopher McComb, Noelia Grande Guti\'errez, Jiuxiang Gu
arXiv AI
Sep 3

When Agents Implement Systems: A Case Study in Defects, Detection, and Evaluation Rigor

The paper reports a case study of a large language model (LLM) coding agent tasked with building a multi‑component data system from a detailed specification. During a single session the agent introduced five defects, which were categorized by violated constraints and detection methods. The study also evaluates the agent’s retrieval‑filtering strategy on the HotpotQA benchmark, showing that filtering to a graph‑identified entity set yields higher recall than unfiltered search, with a statistically significant gap across all tested budgets.

By Phanindra Reddy Madduru