Plausible but Wrong: A case study on Agentic Failures in Astrophysical Workflows

• URL: http://arxiv.org/abs/2604.25345v1
• Date: Tue, 28 Apr 2026 08:01:23 GMT
• Title: Plausible but Wrong: A case study on Agentic Failures in Astrophysical Workflows
• Abstract summary: We evaluate CMBAgent across two workflow paradigms and eighteen astrophysical tasks.<n>In the One-Shot setting, access to domain-specific context yields an approximately 6x performance improvement.<n>In the Deep Research setting, the system frequently exhibits silent failures across stress tests, producing physically inconsistent posteriors without self-diagnosis.
• Score: 10.078675569996992
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Agentic AI systems are increasingly being integrated into scientific workflows, yet their behavior under realistic conditions remains insufficiently understood. We evaluate CMBAgent across two workflow paradigms and eighteen astrophysical tasks. In the One-Shot setting, access to domain-specific context yields an approximately ~6x performance improvement (0.85 vs. ~0 without context), with the primary failure mode being silent incorrect computation - syntactically valid code that produces plausible but inaccurate results. In the Deep Research setting, the system frequently exhibits silent failures across stress tests, producing physically inconsistent posteriors without self-diagnosis. Overall, performance is strong on well-specified tasks but degrades on problems designed to probe reasoning limits, often without visible error signals. These findings highlight that the most concerning failure mode in agentic scientific workflows is not overt failure, but confident generation of incorrect results. We release our evaluation framework to facilitate systematic reliability analysis of scientific AI agents.

Related papers

• Understanding Online Failure Prediction in Linux Through Complementary Multi-View Explainability [0.0]
  This paper reports a practical experience building and evaluating an explainable Online Failure Prediction pipeline for Linux OSs.<n>We combine consensus-based feature selection for detection with temporal onset analysis, subsystemlevel causal analysis, and complementary diagnostic mechanisms to support failure interpretation.
  arXiv  Detail & Related papers  (2026-08-01T13:12:59Z)
• Can Large Language Models Generate Observability-Aware Code? [51.30941152293301]
  We present a systematic study of observability in agent-generated systems.<n>Our results reveal a consistent gap between diagnostic semantics at the source level and fault signals at runtime.<n>We introduce an observability-oriented skill, which can serve as a guidance to improve both diagnostic semantics and fault-signal exposure.
  arXiv  Detail & Related papers  (2026-07-07T03:20:52Z)
• One Reflection Is Not Enough: Self-Correcting Autonomous Research via Multi-Hypothesis Failure Attribution [85.65263343588026]
  We propose SAGE, a Self-correcting, Autonomous, Grounded Experimenter.<n>Its core mechanism, Multi-Hypothesis Failure Attribution (MHFA), treats recovery as a structured causal diagnosis.<n>On a 12-topic, 5-domain benchmark, SAGE increases metrics-bearing outputs from 42% to 92% over a baseline reflection.
  arXiv  Detail & Related papers  (2026-06-30T10:54:16Z)
• MDGYM: Benchmarking AI Agents on Molecular Simulations [25.37963706902467]
  We introduce MDGYM, a benchmark of 169 expert-curated MD simulations spanning LAMMPS and GROMACS.<n>We evaluate three agentic frameworks -- Claude Code, Codex, and OpenHands -- with four LLMs, and find that all perform poorly.<n>These failure modes are qualitatively distinct from those observed in general software engineering benchmarks, indicating that fluent code generation does not transfer to grounded physical reasoning.
  arXiv  Detail & Related papers  (2026-05-09T13:21:51Z)
• LiveFMBench: Unveiling the Power and Limits of Agentic Workflows in Specification Generation [75.05397479715576]
  Large language models (LLMs) and agents have shown promising progress, but their true capabilities and failure modes remain unclear.<n>We present the first systematic and contamination-aware study of LLM- and agent-based formal specification generation for C programs.
  arXiv  Detail & Related papers  (2026-05-02T11:31:33Z)
• Can Coding Agents Reproduce Findings in Computational Materials Science? [49.254975563645786]
  We present AutoMat, a benchmark for evaluating large language models' ability to reproduce scientific claims.<n>By working closely with subject matter experts, we curate a set of claims from real materials science papers to test whether coding agents can recover and execute the end-to-end workflow.<n>Results show that current LLM-based agents obtain low overall success rates on AutoMat, with the best-performing setting achieving a success rate of only 54.1%.
  arXiv  Detail & Related papers  (2026-05-01T17:42:12Z)
• No Test Cases, No Problem: Distillation-Driven Code Generation for Scientific Workflows [5.470408942595905]
  We introduce MOSAIC, a training-free multi-agent framework for scientific code generation without I/O supervision.<n>Instead of execution feedback, MOSAIC employs a student-teacher knowledge distillation framework that grounds generation through domain-specific examples and structured problem decomposition.<n>Experiments on the SciCode benchmark show that MOSAIC improves accuracy, executability, and numerical precision over existing approaches while relying on lightweight models.
  arXiv  Detail & Related papers  (2026-04-25T02:01:54Z)
• Taming Actor-Observer Asymmetry in Agents via Dialectical Alignment [59.536125286960186]
  Multi-agent frameworks assigning specialized roles are increasingly adopted to enable self-reflection and mutual auditing.<n>We find it simultaneously induces a human-like cognitive bias known as Actor-Observer Asymmetry.<n>ReTAS guides agents to synthesize conflicting viewpoints into an objective consensus.
  arXiv  Detail & Related papers  (2026-04-21T15:05:58Z)
• AI scientists produce results without reasoning scientifically [3.100302590436282]
  Large language model (LLM)-based systems are increasingly deployed to conduct scientific research autonomously.<n>Here, we evaluate LLM-based scientific agents across eight domains, spanning execution to hypothesis-driven inquiry.
  arXiv  Detail & Related papers  (2026-04-20T20:23:42Z)
• OccuBench: Evaluating AI Agents on Real-World Professional Tasks via Language Environment Simulation [57.505743202759646]
  OccuBench is a benchmark covering 100 real-world professional task scenarios across 10 industry categories and 65 specialized domains.<n>Our multi-agent synthesis pipeline automatically produces evaluation instances with guaranteed solvability, calibrated difficulty, and document-grounded diversity.
  arXiv  Detail & Related papers  (2026-04-13T00:27:32Z)
• Dynamic analysis enhances issue resolution [53.50448142467294]
  DAIRA (Dynamic Analysis-enhanced Issue Resolution Agent) is an automated repair framework that embeds dynamic analysis into the agent's reasoning cycle.<n> Driven by a Test Tracing-Driven methodology, DAIRA utilizes lightweight monitors to extract critical runtime data.<n>When powered by Gemini 3 Flash Preview, DAIRA establishes a new state-of-the-art (SOTA) performance, achieving a 79.4% resolution rate on the SWE-bench Verified dataset.
  arXiv  Detail & Related papers  (2026-03-23T14:48:54Z)
• Why Do AI Agents Systematically Fail at Cloud Root Cause Analysis? [1.0966260566122241]
  Failures in large-scale cloud systems incur substantial financial losses.<n>Recent efforts leverage Large Language Model (LLM) agents to automate Root Cause Analysis (RCA)<n>This paper presents a process level failure analysis of LLM-based RCA agents.
  arXiv  Detail & Related papers  (2026-02-10T16:14:05Z)
• When Benign Inputs Lead to Severe Harms: Eliciting Unsafe Unintended Behaviors of Computer-Use Agents [90.05202259420138]
  Unintended computer-use agents can deviate from expected outcomes even under benign input contexts.<n>We introduce the first conceptual and methodological framework for unintended CUA behaviors.<n>We propose AutoElicit: an agentic framework that iteratively perturbs benign instructions using CUA execution feedback.
  arXiv  Detail & Related papers  (2026-02-09T03:20:11Z)
• BadScientist: Can a Research Agent Write Convincing but Unsound Papers that Fool LLM Reviewers? [21.78901120638025]
  We investigate whether fabrication-oriented paper generation agents can deceive multi-model LLM review systems.<n>Our generator employs presentation-manipulation strategies requiring no real experiments.<n>Despite provably sound aggregation mathematics, integrity checking systematically fails.
  arXiv  Detail & Related papers  (2025-10-20T18:37:11Z)
• Uncovering Systemic and Environment Errors in Autonomous Systems Using Differential Testing [9.625308787676286]
  We introduce AIProbe, a novel black-box testing technique that applies differential testing to attribute undesirable agent behaviors.<n>AIProbe identifies whether failures are due to errors in the agent's model or policy, or due to unsolvable task conditions.<n>Our evaluation shows that AIProbe significantly outperforms state-of-the-art techniques in detecting both total and unique errors.
  arXiv  Detail & Related papers  (2025-07-05T02:50:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.