CORTEXA
← Browse
arxivcs.SEcs.AI2026-07-19

Where Does Agent Reliability Come From? A Cross-Benchmark Decomposition of Verification Loops, Specialist Models, and Scaffolding in a Production Enterprise Agent

Arunabh Dastidar

Multi-step enterprise agent tasks fail in a characteristic way: single-pass inference has no checkpoint between deciding an answer and committing to it. We study one production system (Leni) whose architecture installs such checkpoints: verification loops (execute, observe, compare, correct) staffed by lightweight task-specialized post-trained models. We evaluate the unmodified production configuration on three public benchmarks stressing distinct failure modes: SpreadsheetBench Verified (silent computation error), BullshitBench v2 (premise confabulation), and the GAIA validation split (cascade error over long tool chains). The full system improves over its frontier base model by +11.0 percentage points on SpreadsheetBench (91.25% vs 80.25%, n=400, p<0.001), +7 to +10 percentage points on BullshitBench (98% vs 91%, n=100), and roughly +15 points on GAIA validation (75.2% pass@1, n=165; 83.0% best-of-k). Our central contribution is a decomposition of that uplift: most of it comes from scaffolding, routing, and specialist models rather than from the verification step itself, whose isolated contribution is small (+1.5 points) but concentrated at the top of the score distribution, where it converts otherwise-failing tasks. We instrument the loop end-to-end, yielding an empirical verifier confusion matrix (catch rate about 0.20, fix rate 0.75, no false-alarm regressions) that grounds a compounding-reliability model. Specialist-swap ablations suggest that the loop's value depends on who observes it: replacing the small trained verifier with the generating frontier model eliminates most rescues. A valid-premise control shows zero over-rejections in 100 expert-level questions.

View free PDFSource page

Related papers

arxivcs.CLcs.AIcs.HCcs.MAcs.SE2026-07-15

DevicesWorld: Benchmarking Cross-Device Agents in Heterogeneous Environments

Huatao Li, Xinwei Geng, Yuheng Wang, Yutong Li, Runde Yang, Hantao Chen, et al.

LLM-based agents have rapidly improved at operating individual digital environments such as mobile applications, desktop systems, and smart homes. However, real-world user goals often span multiple devices: information may come from a phone, be processed on a desktop, and the res…

View free PDFSource page
arxivcs.SEcs.AI2026-07-01

Are Performance-Optimization Benchmarks Reliably Measuring Coding Agents?

Zhi Chen, Zhensu Sun, Yuling Shi, David Lo, Lingxiao Jiang

Repository-level performance-optimization benchmarks such as GSO, SWE-Perf and SWE-fficiency evaluate coding agents by applying patches to real repositories and comparing runtime against unoptimized baselines and official reference patches. Their leaderboard scores are increasing…

View free PDFSource page
arxivcs.SEcs.AIcs.LG2026-07-04

Don't Blame the Large Language Model: How Agent Harness Evolution Shapes Coding Agent Quality

Oussama Ben Sghaier, Hao Li, Bram Adams, Ahmed E. Hassan

Coding agents, autonomous systems that use large language models (LLMs) to resolve software engineering tasks, rely on agent harness: a middleware layer in between a developer and a large language model that orchestrates system prompts, tool execution, context management, and ite…

View free PDFSource page
arxivcs.SEcs.AI2026-07-04

The "I Don't Know" Filter: Enhancing Agentic Reliability in Function Calling

Stefan Broecker, Mason del Rosario, Boris Selitser, Thomas Strohmer

The language models that underpin agents have seen a rapid rise in performance on function calling benchmarks. However, the metrics used in the training and evaluation of these models often encourage models to make positive claims even when the answer is uncertain, leading to hal…

View free PDFSource page
arxivcs.SEcs.AI2026-07-16

Beyond Generalist LLMs: Specialist Agentic Systems for Structured Code Workflow Execution

Harris Borman, Herman Wandabwa, Fusun Yu, Sandeepa Kannangara, Justin Liu, Anna Leontjeva, et al.

Large Language Models (LLMs) have accelerated the adoption of software development agents, now widely available as Integrated Development Environment (IDE) extensions and standalone applications. While these agents are typically general-purpose, it remains unclear whether special…

View free PDFSource page